Educating English Language Learners

P1: irk
0521859751pre CB989B/Genesee 0 521 85975 1 November 12, 2005 17:3
This page intentionally left blank
i
P1: irk
Educating English Language Learners: A Synthesis

of Research Evidence
The book provides an extensive review of scientific research on the learning outcomes of
students with limited or no proficiency in English in U.S. schools. Research on students
in kindergarten through grade 12 is reviewed. The primary chapters of the book focus
on these students’ acquisition of oral language skills in English, their development of
literacy (reading and writing) skills in English, instructional issues in teaching literacy,
and achievement in academic domains (i.e., mathematics, science, and reading). The
reviews and analyses of the research are relatively technical with a focus on research
quality, design characteristics, and statistical analyses. The book provides a unique set
of summary tables that give details about each study, including full references, charac-
teristics of the students in the research, assessment tools and procedures, and results.
A concluding chapter summarizes the major issues discussed and makes recommenda-
tions about particular areas that need further research.
Fred Genesee is Professor in the Department of Psychology, McGill University, Montreal.

He has conducted extensive research on second language education, including second
language immersion programs for majority language students and bilingual education
for minority language students. His research on bilingual acquisition focuses on the syn-
tactic and communicative development of bilingual children with typical and impaired
patterns of acquisition and addresses issues related to the capacity of the language
faculty during the period of primary language development.
Kathryn Lindholm-Leary is Professor of Child and Adolescent Development at San Jose

State University where she has taught for seventeen years. Her research interests focus on
understanding the cognitive, language, psychosocial, and societal factors that influence
student achievement, with a particular emphasis on culturally and linguistically diverse
students. Kathryn has worked with dual language education programs for the past
twenty years and during that time has evaluated more than thirty programs and helped
to establish programs in more than fifty school districts in ten states.
William M. Saunders, Ph.D., is a Senior Research Fellow at CSU, Long Beach, at UCLA,
and at LessonLab. He currently directs LessonLab’s school-based programs for improv-
ing teaching, learning, and schooling. He has directed several research programs includ-
ing longitudinal studies of the literacy development of English learners, clinical trials
of discrete instructional components, and prospective studies of school improvement.
Formerly a high school teacher and Director of the Writing Project at the University
of Southern California, Saunders has conducted school improvement and professional
development programs in the Southern California region for the past twenty years. He
is the author of numerous papers and chapters on literacy instruction, school change,
assessment, and English language learners.
Donna Christian is President of the Center for Applied Linguistics (CAL) in Washington,
DC. She has worked with CAL since 1974, focusing on the role of language in education,
including issues of second language learning and dialect diversity. Among her activities,
she has directed a program on two-way bilingual immersion, including a study for the
National Center for Research on Education, Diversity & Excellence (CREDE), funded by
the U.S. Department of Education. She is also a senior advisor to the Heritage Languages
Initiative, the Biliteracy Research Program, and the National Literacy Panel on Language
Minority Children and Youth.
i
P1: irk
ii
P1: irk
Educating English Language Learners

A Synthesis of Research Evidence
FRED GENESEE
McGill University
KATHRYN LINDHOLM-LEARY
San Jose State University
WILLIAM M. SAUNDERS
California State University, Long Beach
DONNA CHRISTIAN
Center for Applied Linguistics
iii
cambridge university press
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo
Cambridge University Press

The Edinburgh Building, Cambridge cb2 2ru, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521859752
© Cambridge University Press 2006
This publication is in copyright. Subject to statutory exception and to the provision of

relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
First published in print format
isbn-13 978-0-511-14648-0 eBook (EBL)

isbn-10 0-511-14648-5 eBook (EBL)
isbn-13 978-0-521-85975-2 hardback

isbn-10 0-521-85975-1 hardback
isbn-13 978-0-521-67699-1
isbn-10 0-521-67699-1
Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
P1: irk
Contents
List of Tables page vii

Preface ix
1. Introduction 1
D. Christian
2. Oral Language 14
W. Saunders and G. O’Brien
3. Literacy: Crosslinguistic and Crossmodal Issues 64
C. Riches and F. Genesee
4. Literacy: Instructional Issues 109
F. Genesee and C. Riches
5. Academic Achievement 176
K. Lindholm-Leary and G. Borsato
6. Conclusions and Future Directions 223
F. Genesee, K. Lindholm-Leary, W. Saunders, and D. Christian
Appendix A: Definitions of Abbreviations in Research Summary Tables 235

Index 239
v
P1: irk
vi
P1: irk
List of Tables
1.1. Characteristics of Program Alternatives for English

Language Learners page 4
2.1. Correlations Between L2 Oral and L2 Reading Measures 18
2.2. Results of Longitudinal and Cross-Sectional Studies of
Oral Language Outcomes 24
A.2.1. Summary of Studies on Language Development 46
A.2.2. Summary of Studies on School Factors 52
A.2.3. Summary of Studies on Non-School Factors 59
A.2.4. Summary of Studies on Assessment 61
A.3.1. Summary of Studies on L1 Oral Proficiency and L2
Literacy 88
A.3.2. Summary of Studies on L2 Oral Proficiency and L2
Literacy 92
A.3.3. Summary of Studies on Component Skills of L2 Literacy
Development 96
A.3.4. Summary of Studies on L1 Literacy and L2 Literacy 101
4.1. Sample Descriptions of Interactive Learning Environments 117
4.2. Sample Descriptions of Process Approaches 122
A.4.1. Summary of Studies on Direct Instruction 149
A.4.2. Summary of Studies on Interactive Instruction 155
A.4.3. Summary of Studies on Process Instruction 162
A.4.4. Summary of Studies on Language of Instruction 166
A.4.5. Summary of Studies of Family Influences 171
A.5.1. Comparative Evaluation Studies 212
A.5.2. Descriptive Studies of Two-Way Immersion Programs 220
vii
P1: irk
viii
P1: irk
Preface
This book grew out of the work of the Center for Research on Educa-
tion, Diversity & Excellence (CREDE). CREDE was a center funded by
the U.S. government1 to conduct research, generate knowledge, and pro-
vide services to improve the education of students whose ability to reach
their potential is challenged by language or cultural barriers, race, geo-
graphic location, or poverty. From 1996 to 2001, CREDE comprised thirty-
one research projects around the country that sought to extend knowledge
about the education of the diverse students who make up the U.S. school
population, from kindergarten through grade 12. These research projects
were organized around six themes that are integral to the education of
diverse students: language learning and academic achievement; profes-
sional development; family, peers, and community; instruction in context;
integrated school reform; and assessment. Researchers working on each
theme gathered data and tested curriculum models in wide-ranging set-
tings and with diverse student populations – from classrooms with pre-
dominantly Zuni-speaking students in New Mexico to inner-city schools in
Florida to California elementary schools with large populations of native
Spanish-speaking students.
Following the completion of the first phase of research in 2001, CREDE
researchers extended the knowledge base that can be used to improve
the education of diverse students by carrying out systematic, thorough,
and critical reviews of research related to the themes. Seven synthesis
teams were created, each involving researchers, practitioners, and policy
experts, to survey and critique the available research on a theme and make
1 This work was supported under the Education Research and Development Program,
PR/Award R306A60001, the Center for Research on Education, Diversity & Excellence,
as administered by the Institute of Education Sciences (IES) of the U.S. Department of
Education. The contents, findings, and opinions expressed in this volume are those of the
authors and do not necessarily represent the positions or policies of IES or the U.S. ED.
ix
P1: irk
x Preface
recommendations for future research agendas. The chapters in this volume

report on the work of the team charged with reviewing research on the lan-
guage and academic development of students who come to school with
no proficiency or limited proficiency in English; that is, English language
learners (ELLs). The volume reviews and summarizes scientific research
on three fundamental aspects of the education of ELL students: their oral
language development, their literacy development, and their academic
development.
The team members, in addition to the authors, who guided the syn-
thesis work consisted of Diane August (Center for Applied Linguistics),
Gil Cuevas (University of Miami), Else Hamayan (Illinois Resource
Center), Liliana Minaya-Rowe (University of Connecticut), Mary Ramirez
(Pennsylvania Department of Education), Noni Reis (San Jose State Univer-
sity), Charlene Rivera (The George Washington University), Deborah Short
(Center for Applied Linguistics), and Sau-Lim Tsang (ARC Associates).
We also recognize and appreciate the assistance provided by individuals
who co-authored specific chapters of this volume: Graciela Borsato, Gisela
O’Brien, and Caroline Riches. Their assistance was critical in the success-
ful completion of this work. Finally, we are grateful for the leadership
of Roland Tharp, the director of the Center for Research on Education,
Diversity & Excellence (CREDE) for developing and encouraging the syn-
thesis process, and we thank Yolanda Padrón, assistant director of CREDE,
for her support during the project.
P1: PJU
0521859751c01 CB989B/Genesee 0 521 85975 1 November 11, 2005 20:56
Introduction
Donna Christian
This volume synthesizes research on the relationships among oral lan-

guage, literacy, and academic achievement for English language learners
(ELLs) in the United States, from pre-Kindergarten through Grade 12. It
explores how these findings have been applied in school and classroom set-
tings and recommends areas of focus for future studies in order to improve
education for these students.
Why is it important to assess what we know about the education of
ELLs? The most basic reason, of course, is that we seek to provide, for
ALL students, a high quality education that takes into account their indi-
vidual strengths and needs. The level of academic achievement for stu-
dents with limited proficiency in English in the United States has lagged
significantly behind that of native English speakers. One congressionally
mandated study reported that ELLs receive lower grades, are judged by
their teachers to have lower academic abilities, and score below their class-
mates on standardized tests of reading and mathematics (Moss and Puma,
1995). According to a compilation of reports from forty-one state educa-
tion agencies, only 18.7 percent of students classified as limited English
proficient (LEP) met the state norm for reading in English (Kindler, 2002).
Furthermore, students from language minority backgrounds have higher
dropout rates and are more frequently placed in lower ability groups and
academic tracks than language majority students (Bennici and Strang, 1995;
President’s Advisory Commission on Educational Excellence for Hispanic
Americans, 2003; Ruiz-de-Velasco and Fix, 2000).
These educational facts intersect with the demographic facts to
strengthen the rationale for this research synthesis. Across the nation, the
number of students from non-English-speaking backgrounds has risen dra-
matically. They represent the fastest growing segment of the student popu-
lation in the United States by a wide margin. From the 1991–92 school year
through 2001–02, the number of identified students with limited English
proficiency in public schools (K-12) grew 95 percent while total enrollment
1
P1: PJU
2 Donna Christian
increased by only 12 percent. In 2001–02, over 4.7 million school-aged chil-

dren were identified as LEP (the term used in the survey), almost 10 percent
of the K-12 public school student population (National Clearinghouse for
English Language Acquisition, 2003). These students speak over 400 lan-
guages, but nearly 80 percent are native Spanish speakers. Of the remaining
20 percent, the largest language groups are Vietnamese (2 percent), Hmong
(1.6 percent), Cantonese (1 percent), and Korean (1 percent) (Kindler, 2002).
ELL students come to U.S. schools with many resources, including lin-
guistic resources in their native language. However, they enter U.S. schools
with a wide range of language proficiencies (in English and in other lan-
guages) and of subject-matter knowledge. They differ in educational back-
ground, expectations of schooling, socioeconomic status, age of arrival in
the United States, and personal experiences coming to and living in the
United States.
Among ELLs who are immigrants, some have strong academic prepa-
ration. They are at or above equivalent grade levels in the school curricula
and are literate in their native language. Other immigrant students enter
U.S. schools with limited formal schooling – perhaps due to war or the
isolated location of their home. Ruiz-de-Velasco and Fix (2000) found that
20 percent of all ELLs at the high school level and 12 percent of ELLs at the
middle school level have missed two or more years of schooling since age
six. Among Hispanic students aged 15–17, more than one third are enrolled
below grade level (Jamieson, Curry, and Martinez, 2001). These students
are not literate in their native language; they have never taken a standard-
ized test. They have significant gaps in their educational backgrounds,
lack knowledge in specific subject areas, and often need additional time to
become accustomed to school routines and expectations.
Students who have been raised in the United States but speak a language
other than English at home may or may not be literate in their home lan-
guage. Some have strong oral English skills; others do not. Most of the U.S.-
born ELLs begin their education in the U.S. public schools. There they must
learn basic skills, including initial literacy. They may have some prepara-
tion for schooling from participation in pre-school programs, but U.S.-born
ELLs have as much diversity in backgrounds as older immigrant students.
Although English language proficiency is a critical factor in educational
success in this country, there are many other factors that can put students at
risk for educational failure, and a number of these factors tend to co-occur
with limited English proficiency (Garcia, 1997; Tharp, 1997). These include
economic circumstances, race, educational environment, geographic loca-
tion, immigration status, health, and many others. Although research on
all of those factors is relevant to improving the education of English lan-
guage learners in general, this synthesis focuses on such factors only as
they relate to oral language, literacy, and academic achievement.
P1: PJU
Introduction 3
The increasing number of students for whom English is an additional

language is particularly significant in light of educational reform that calls
for high standards and strong accountability for schools and students.
Although many states exempt ELLs from state-mandated tests for a period
of time, the amount of time may be insufficient for some ELLs to acquire and
apply academic English. For example, an immigrant student who enters
high school with no English proficiency may be expected to pass tests
for graduation in mathematics, biology, English language arts, and other
subjects, after three (or fewer) years of U.S. schooling.
Federal programs in the United States have also increased the emphasis
on accountability. For example, No Child Left Behind, the 2001 reauthoriza-
tion of the Elementary and Secondary Education Act, calls for annual tests
of reading and mathematics for all students in Grades 3–12 (in schools
receiving federal funds under the law) and deliberately includes ELLs in
state accountability systems. Although schools may exempt ELLs from
achievement testing in English for up to three years, they must assess
English language proficiency annually (with no exemption period).
Improved education is key to improving performance for ELLs on these
tests, and research results can inform such improvements.
program alternatives for ells

English language learners face the dual challenges of mastering English
and acquiring the academic skills and knowledge deemed essential for a
sound education and a productive future life. Schools face the challenge of
designing programs to help ELLs achieve these goals. As mentioned earlier,
ELL students embody diversity at many levels, including their socioeco-
nomic status, the types of neighborhoods in which they live, the varieties
of English and/or other languages they speak, and their cultural back-
grounds. The challenge is magnified by the fact that these students are
entering U.S. schools at every grade level and at various times during
the academic year. Students who enter at the elementary level, of course,
have the advantage of more time to acquire the language and academic
skills they need (compared to ELLs who enter at the secondary level). The
availability of time, however, does not lessen the need for appropriate and
challenging instruction, from the very beginning, through the first and/or
second language.
Genesee (1999) discusses a set of program alternatives that may meet
the diverse and complex needs of ELLs (see Table 1.1). Some of them incor-
porate content instruction in the native language. Two-way immersion pro-
grams serve ELLs who speak a common native language along with native
English speakers (Howard and Christian, 2002). For both groups of stu-
dents, the goals are to develop high levels of first and second language
P1: PJU
4
0521859751c01
table 1.1. Characteristics of Program Alternatives for English Language Learners (Adapted with permission from Genesee, 1999)
Structured
Two-Way Developmental Transitional English Sheltered
Immersion Bilingual Bilingual Newcomer ESL Immersion Instruction
Language Goals Bilingualism Bilingualism English English English English Proficiency in
CB989B/Genesee
proficiency proficiency proficiency proficiency academic

English
Cultural Goals Integrate into Integrate into Integrate into Integrate into Integrate into Integrate into Integrate into
mainstream mainstream mainstream mainstream mainstream mainstream mainstream
American American American American American American American
culture & culture & culture culture culture culture culture
0 521 85975 1
maintain/ maintain/
appreciate ELL appreciate ELL
home culture home culture
Language(s) L1 of ELLs and L1 of ELLs and L1 of ELLs and English (some English English English
of Instruction English English English programs use L1)
Students Both native & Non-native Non-native No/limited English Non-native English No/limited Non-native
non-native speakers of speakers of Low level literacy speakers with English; speakers of
(with same L1) English with English with Recent arrivals; various levels of varied L1 & English;
speakers of same L1; same L1; varied L1 & Eng proficiency; cultural varied L1 &
English varied varied cultural varied L1 & backgrounds cultural
cultural cultural backgrounds cultural backgrounds
backgrounds backgrounds backgrounds
Grades Served K-12 Primarily Primarily K-12; many at K-12 Primarily K-12
elementary elementary middle & high elementary
school levels
November 11, 2005
Typical Length 5–12 years 5–12 years 2–4 years 1–3 semesters 1–3 years 1 year 1–3 years
of Participation
20:56
P1: PJU
Introduction 5
proficiency, academic development, and crosscultural understanding. All

students experience an additive bilingual environment (one in which both
languages are valued and developed), and academic content is learned
through two languages. These are typically full K-6 or K-12 instructional
programs.
Developmental bilingual programs provide a similar additive bilingual
environment, with a goal of high levels of proficiency in two languages, but
the students served are primarily or solely ELLs. This model, also referred
to as “late-exit” or maintenance bilingual education, uses both English
and the students’ native language for academic instruction and promotes
sustained development of the first language as well as English. Students
generally participate in these programs for five to six years.
In transitional bilingual programs (also known as “early-exit” bilingual
education), academic instruction in the students’ native language is pro-
vided while they learn English (to varying extents and for varying lengths
of time) through ESL classes. As their English proficiency develops, stu-
dents are exited from the program and placed in all English, mainstream
classes, typically after one to three years.
Newcomer programs are specially designed programs for recent arrivals
to the United States, who have no or low English proficiency and often lim-
ited literacy in their native language (Short and Boyson, 2004). The goal is
to accelerate their acquisition of language and academic skills and to orient
them to the United States and U.S. schools. Students typically participate
in such programs for one to one and one-half years. Although newcomer
programs exist in elementary schools, they are more prevalent at the sec-
ondary level. Some programs follow a bilingual approach; others focus
on sheltered instruction in English (see the later discussion of sheltered
instruction).
Other program models offer primarily English instruction to ELLs. This
choice is often made when ELLs in a school come from many different
language backgrounds. In English as a Second Language (ESL) programs
(also known as English language development [ELD] programs), carefully
articulated, developmentally appropriate English language instruction is
designed to meet the needs of students at various levels of English pro-
ficiency. ELLs may receive content instruction from other sources while
they participate in the ESL program, or they may be in self-contained class-
rooms. Students generally participate in ESL programs for one to five years,
depending on their initial level of proficiency and rate of progress. Students
often benefit greatly when programs provide various kinds of support after
they have moved fully into English mainstream classes, to give targeted
assistance as needed. Structured English immersion is a form of ESL pro-
gram taught in self-contained classrooms where most instruction is pro-
vided in English, though use of the student’s native language is possible.
P1: PJU
6 Donna Christian
The core curriculum includes English language development (ELD), and

content area instruction is taught using special techniques for second lan-
guage learners (Baker, 1998). This program type has become most well
known as the approach prescribed by state referenda (e.g., Proposition 227
in California) that restrict the use of bilingual education programs.
Another ESL-oriented program model is sheltered instruction, which is
often found in school systems with ELLs from multiple language back-
grounds. Sheltered programs offer ELLs a comprehensive, articulated pro-
gram where the regular grade-level, core content courses are taught in
English through instructional strategies that make the content concepts
accessible to ELLs and that promote the development of academic English
(Short and Echevarria, 1999). Sheltered instruction teachers should have
ESL or bilingual education training in addition to training in the content
area, and they often form a school team or learning community. Most shel-
tered instruction programs are designed to meet all the requirements for
credit toward grade-level promotion or graduation. Students remain in
them for two to three years. The term sheltered instruction may also be used
to describe pedagogy rather than a program design. Sheltered instruction
practices and individual sheltered instruction courses can be and often are
implemented in conjunction with other program alternatives.
These program models differ in certain dimensions. Some set a goal
of bilingualism for language development (two-way immersion, devel-
opmental bilingual), while others emphasize proficiency in English (ESL,
sheltered instruction). The characteristics of the appropriate student pop-
ulation vary, particularly in terms of the homogeneity of native language
backgrounds. The typical length of student participation also differs, with
some programs being intended as short-term or transitional (one to four
years) and others longer in duration (six or more years). The resources
required vary from model to model, in terms of teacher qualifications (lan-
guage skills and professional preparation), curricula and materials (how
extensive bilingual offerings need to be), and so on.
methodology for the review

Our synthesis is based on a systematic review of the research literature. The
goal was to be as comprehensive as possible in terms of the student popu-
lation, to include ELLs from pre-Kindergarten through Grade 12 of diverse
language backgrounds in educational programs in the United States. Given
the demographic characteristics of the United States, however, most of the
published research on ELLs focuses on low-income native Spanish speak-
ers, and the largest number of studies involve elementary school-aged
students. This will undoubtedly limit the generalizability of the results to
other language and age groups, but it also highlights areas where future
research is clearly needed.
P1: PJU
Introduction 7
We convened a thirteen-member team of researchers knowledgeable

about the education of ELLs to conduct the synthesis of research in this
area (see Preface for a list of team members). The team met three times
during the two-year project period to set the parameters for the synthesis,
review the findings of literature searches, and review drafts of sections of
the synthesis.
The synthesis was conducted in three phases, parallel to the team meet-
ings. In the first phase, the team defined the scope of the synthesis and the
research to be reviewed. In the second phase, we conducted searches of
the literature according to the defined parameters and evaluated the doc-
uments identified in the searches for relevance and quality. Finally, in the
third phase, we synthesized the relevant research that met basic quality
criteria and formulated our conclusions.
Phase 1: Inclusion Parameters

As mentioned previously, the focus of the research synthesis is the devel-
opment of oral language, literacy, and academic achievement for ELLs
in a variety of alternative programs, including English mainstream class-
rooms. The synthesis examined only English learners and did not consider
research on ethnic minority or immigrant students except as the samples
and results specifically address ELLs. For the searches of the literature,
the following parameters were set to define which research studies to
include:
r Empirical
r Conducted in the United States
r Published in English
r Focused on oral language development, literacy, and academic achieve-
ment among ELLs, with outcome measures in English
r Focused on pre-K through Grade 12
r Published in the last 20 years (may include seminal works conducted
earlier)
r Published as peer-reviewed journal articles and selected technical
reports (no books, book chapters, or dissertations)
r In the case of literacy, included reading, writing, or reading- or writing-
related outcomes
Phase 2: Literature Searches and Quality Indicators

The synthesis was divided into three parts: oral language, literacy, and aca-
demic achievement. A subgroup of the team reviewed the literature in each
area, considered which articles to include in the synthesis, and compiled
P1: PJU
8 Donna Christian
the research. The subgroups conducted several types of searches. First, we

searched three large databases of language and education materials using
specific key terms – the Education Resources Information Center (ERIC),
Linguistics and Language Behavior Abstracts (LLBA), and PsycInfo. The
key terms for the searches included limited English speaking, academic achieve-
ment (and math, science, social studies), English second language, bilingual edu-
cation, literacy, reading, and oral language proficiency. In addition, team mem-
bers searched a number of education journals by hand (see Appendix for
list of journals searched) and reviewed technical reports from several fed-
erally funded research centers (see Appendix for list). As a result of the
computer and hand searches, over four thousand articles and reports were
considered (at least by title, but most often by abstract).
To facilitate the processing and synthesizing of studies, we developed
a coding system to record pertinent information for each study into a
database. Each entry included the bibliographic citation, type of study,
analytic methods, research questions addressed, information about the
methodology, information about the sample, and the domain of the syn-
thesis that the study addresses (oral language, literacy, and/or academic
achievement).
We reviewed each of the abstracts obtained from the computerized
searches for relevance to the topic and entered those that met the crite-
ria into a database along with the articles and technical reports identified
through hand searching. About five hundred articles and reports were
reviewed at this level. In several instances, articles and reports were rel-
evant to two subtopics. For instance, some studies fit into both literacy
and academic achievement because student outcomes on both reading
and mathematics were reported. Each subgroup reviewed all the studies
relevant to their domain, so some studies were reviewed by more than one
subgroup. We then obtained full texts of the articles, and reviewed their
bibliographies to identify additional resources. When we found relevant
articles in those bibliographies, they were added to the database as well.
Each article was read and annotated, according to a coding framework
for entries into the database. Based on this coding, articles that did not
qualify for inclusion for relevance or quality reasons (discussed next) were
rejected and not included in the synthesis. If two or more articles contained
the same analyses based on the same data, only the more complete one was
included.
The guiding principles for scientific research in education identified
in the National Research Council report on Scientific Research in Edu-
cation (Shavelson and Towne, 2001) formed the basis for the quality
indicators used to examine the articles under consideration. The team
looked for (a) appropriate research design to answer the questions being
posed; (b) research that was well carried out and clearly described; and
P1: PJU
Introduction 9
(c) conclusions that were supported by the evidence presented. Some par-
ticular qualifiers were:
r Careful description of study participants (including age, language
and ethnic background, socioeconomic status, other relevant charac-
teristics);
r Sufficient detail in description of study interventions (including length
and type of treatment) to allow for replicability;
r Description of study methods sufficient to allow judgments about how
instructional fidelity was ensured, where appropriate;
r Full description of testing instruments, data-collection procedures, out-
come measures, and data analytic techniques (all of which are appro-
priate for the goals of the study);
r Empirical outcomes reported;
r Study conclusions and implications clearly and reasonably linked to
data.
The result was a set of articles that could be included in the synthesis for
each topic. The final corpus included in the synthesis contained approxi-
mately two hundred articles and reports.
During the search and evaluation process, studies were coded accord-
ing to information given by the authors of the research reports. Each of the
following chapters includes tables summarizing relevant characteristics of
the studies being synthesized in a given section (e.g., sample characteris-
tics, outcome measures). The descriptions in these tables reflect the terms
used by the authors of the articles, in order to avoid making any infer-
ences about the characteristics (e.g., description of the sample students
as “Hispanic,” “Mexican American,” or “Latino/a”). In addition, the cat-
egories of information provided in the tables vary in some cases across
domains (the chapter topics) in order to suit the research represented (e.g.,
the tables in Chapter 4 include the category “instructional methods” while
others do not). Definitions of abbreviations used in these tables are pro-
vided in Appendix A at the end of the book.
Phase 3: Synthesis of Research

As we reviewed the research studies for relevance and quality, we also
sorted them by themes that captured the features of the research base.
Within each domain (oral language, literacy, and academic achievement),
we grouped together studies that addressed topics like instructional fac-
tors, home/community factors, assessment, and so on. At first, we planned
to examine the research on each domain according to a common set of
dimensions. However, it became clear that studies in each domain clus-
tered in different ways, and we allowed those clusters to emerge from
P1: PJU
10 Donna Christian
the research base. Thus, we developed a framework for characterizing the

corpus of studies for each domain. In some cases, that corpus of studies
was very eclectic and some subthemes clustered better than others. When
we found several studies that addressed the same questions, we could
draw stronger generalizations. When we found only one or two studies
that looked at an issue, we could not generalize. In the discussion, we pro-
vide different levels of detail on different studies, depending on how well
they fit with a group of studies to address certain questions. As a result,
we may only briefly mention some studies in the database because they
do not fit well into a cluster with other studies.
Once we identified the themes for each topic area, we reviewed the
studies in each theme as a group and synthesized them. Various team
members took the lead in drafting the synthesis chapters and then the
entire team reviewed them. At the final meeting, team members revisited
and revised the themes within domains to better fit the research base that
was found, and they identified strengths and gaps in the research base.
As the team finalized the syntheses of research by domain, it developed
recommendations on future directions for research in this area.
organization of the volume

The organization of the research synthesis parallels the major research areas
that we explored: oral language, literacy, and academic achievement. Lit-
eracy is divided into two chapters, one dealing with crosslinguistic and
crossmodal issues in literacy development and one on instructional issues.
Different members of the team developed these sections, as authorship
indicates, but worked closely together to ensure comparable methods and
complementary scope. The full team reviewed all the sections. As men-
tioned earlier, the scopes of the three domains overlap, so some studies
appear in more than one chapter. The final chapter offers conclusions that
may be derived from the synthesis as well as recommendations for future
research.
Before moving on to the synthesis itself, a note about terminology and
labels may be useful. This is an area of considerable complexity in studies
that involve students who come from homes where a language other than
English is spoken. Research in the field suffers from inconsistency in def-
initions of categories into which the students may be grouped and incon-
sistent application of definitions to student populations by researchers and
practitioners.
In this volume, “English language learner (ELL)” is used as the term
for students who first learn a language other than English in their home
and community (U.S.-born or immigrant) and then learn English as a new
language. When they enter school in the United States, they may or may
not have some knowledge of English, but they are not yet fully proficient.
P1: PJU
Introduction 11
In the past, a more common label for these students was “limited English
proficient” or “LEP.” This term has a legislative history in the federal
government and remains the one in use in federal-policy contexts. Detailed
legal definitions are provided in such legislation as the No Child Left Behind
Act of 2001 in order to specify terms for eligibility for services and applica-
bility of various requirements. Other terms often used include non-native
English speaker, language minority student, ESL student, or bilingual stu-
dent. In reporting the results of studies here, our best attempts were made
to determine how the subject populations were characterized; however,
this remains an area of concern in interpreting the research.
For native speakers of English (who may be compared with ELLs), the
label “English-only” (“EO”) is often used, signifying the monolingual lan-
guage skills possessed by these students. Another convention that will
be maintained in the chapters that follow is the use of “L1” to refer to an
individual’s native or home language and “L2” for the second (or later) lan-
guage. Thus, a native Spanish-speaking student who is learning English
could be described as having Spanish L1 and English L2.
References
Baker, K. 1998. Structured English immersion: Breakthrough in teaching limited-
English-proficient students, Phi Delta Kappan (November).
Bennici, F. J., and Strang, E. W. 1995. An analysis of language minority and limited
English proficient students from NELS 1988. Report to the Office of Bilingual
Education and Minority Languages Affairs, U.S. Department of Education,
August 1995.
Garcia, G. N. 1997. Placing a face on every child and youth. Talking Leaves 2(1): 3,
7. Fall, 1997. (Newsletter of the Center for Research on Education, Diversity &
Excellence.)
Genesee, F. (ed.). 1999. Program alternatives for linguistically diverse students.
Educational Practice Report No. 1. Santa Cruz, CA, and Washington,
DC: Center for Research on Education, Diversity & Excellence. Available:
http://www.cal.org/crede/pubs/edpractice/EPR1.htm.
Howard, E., and Christian, D. 2002. Two-way immersion 101: Designing and imple-
menting a two-way immersion program at the elementary level. Educational Practice
Report No. 9. Santa Cruz, CA, and Washington, DC: Center for Research on
Education, Diversity & Excellence.
Jamieson, A., Curry, A., and Martinez, G. 2001. School enrollment in the United
States – Social and economic characteristics of students. Current Population
Reports, P20–533. Washington, DC: U.S. Government Printing Office. Available:
http://www.census.gov/prod/www/abs/school.html.
Kindler, A. L. 2002. Survey of the states’ limited English proficient students and avail-
able educational programs and services: 2000–2001 summary report. Washington,
DC: National Clearinghouse for English Language Acquisition & Language
Instruction Educational Programs.
P1: PJU
12 Donna Christian
Moss, M., and Puma, M. 1995. Prospects: The congressionally mandated study of edu-
cational growth and opportunity. First year report on language minority and limited
English proficient students. Washington, DC: National Clearinghouse for Bilin-
gual Education.
National Clearinghouse for English Language Acquisition. 2003. The growing num-
bers of limited English proficient students 1991–92–2001–02. Retrieved July 31,
2003, from http://www.ncela.gwu.edu/states/stateposter.pdf.
President’s Advisory Commission on Educational Excellence for Hispanic Amer-
icans. 2003. From risk to opportunity: Fulfilling the educational needs of Hispanic
Americans in the 21st century. Washington, DC.
Ruiz-de-Velasco, J., and Fix, M. 2000. Overlooked & underserved: Immigrant students
in U.S. secondary schools. Washington, DC: The Urban Institute.
Shavelson, R. J., and Towne, L. (eds.). 2001. Scientific research in education. Washing-
ton, DC: National Academy Press.
Short, D. J., and Boyson, B. 2004. Creating access: Language and academic programs for
secondary school newcomers. McHenry, IL, and Washington, DC: Delta Systems
and Center for Applied Linguistics.
Short, D. J., and Echevarria, J. 1999. The Sheltered Instruction Observation Protocol:
A tool for teacher-researcher collaboration and professional development. Educa-
tional Practice Report No. 3. Santa Cruz, CA, and Washington, DC: Center
for Research on Education, Diversity & Excellence.
Tharp, R. 1997. From at-risk to excellence: Research, theory, and principles for prac-
tice. Research Report No. 1. Santa Cruz, CA, and Washington, DC: Center for
Research on Education, Diversity & Excellence.
appendix to chapter 1
Journals Searched
Annual Review of Applied Linguistics
Applied Linguistics
Applied Psycholinguistics
Bilingual Research Journal
Educational Researcher
Elementary School Journal
Harvard Education Review
International Journal of Bilingual Education and Bilingualism
Journal of Education
Journal of Education of Students Placed at Risk
Journal of Educational Issues of Language Minority Students
Journal of Learning Disabilities
Journal of Reading Behavior
Language and Education
Language Learning
Modern Language Journal
P1: PJU
Introduction 13
NABE Journal
Peabody Journal of Education
Phi Delta Kappan
Studies in Second Language Acquisition
TESOL Quarterly
Federal Research Centers Technical Reports Searched

Center for Research on Education, Diversity & Excellence
National Center for Research on Cultural Diversity and Second Lan-
guage Learning
Center for Language Education and Research
Center for Research on the Education of Students Placed At Risk
Center for Research on Evaluation, Standards, and Student Testing
P1: PJU
Oral Language
William M. Saunders and Gisela O’Brien
introduction
For English Language Learners (ELLs) in U.S. schools, developing profi-
ciency in oral English is essential for academic and future professional and
personal success. Developing proficiency in oral English involves acquir-
ing vocabulary, gaining control over grammar, and developing an under-
standing of the subtle semantics of English. At the same time, acquiring
proficiency in English involves learning how to use the language to inter-
act successfully with other speakers of the language. Oral interactions can
vary considerably from exchanging greetings to initiating and sustaining
conversations to negotiating collaborative tasks to giving and/or receiving
directions to telling or listening to stories to delivering or comprehending
lectures.
While the design of programs for ELLs varies in terms of the use of
L1, many programs (that is, ESL pull-out, English immersion, transitional
bilingual, developmental bilingual, and two-way immersion) recommend
daily oral English language instruction until students achieve at least a min-
imum level of proficiency in English (see Genesee, 1999, for a description
of alternative programs). Moreover, while there are different theoretical
views about the minimum level of English oral proficiency necessary for
successful participation in classrooms with English reading, writing, and
content area instruction (Baker, 1998; Cummins, 1979; Fitzgerald, 1995;
Krashen, 1996), there is no controversy about the fundamental importance
of English oral language development as part of the larger enterprise of
educating ELLs.
Despite the centrality afforded English oral language development in
both theory and practice, the empirical literature on oral language develop-
ment in ELLs is small. Our search for studies on the English oral language
development of ELLs turned up approximately one fourth of the number
of studies recovered for literacy. Of the approximately 150 studies on oral
14
P1: PJU
Oral Language 15
language development, fewer than two thirds reported actual oral lan-
guage outcomes, and fewer than one third reported oral language out-
comes and also met criteria for relevance and methodological adequacy.
Moreover, the studies that were retained vary considerably; some mea-
sured general oral language proficiency; others measured discrete elements
of oral language proficiency (e.g., vocabulary); and yet others measured
language choice and use. Such variation makes synthesis and generaliz-
ability difficult.
We clustered studies that met our criteria topically. In some topic clus-
ters, we found a sufficient number of studies to warrant firm conclusions.
For most topics, however, the small number of studies allows for only
qualified conclusions that best serve as hypotheses for future research. We
thoroughly reviewed and analyzed the studies that were retained, looking
for every opportunity to utilize data, datasets, and findings to address rel-
evant topics. Given the paucity of research on oral language development,
we thought it best to retain a topic, even if it was addressed by only two
or three studies, in order to encourage and inform future research on that
topic. The following review is organized according to these topics:
1. Language Development
2. School Factors
3. Non-school Factors
4. Assessment
language development
In this section, we synthesize research findings that establish some of the
major characteristics of English oral language development among ELLs
in U.S. schools (see Table A.2.1 for studies reviewed in this section). Topics
include (1) Specific Features of L2 Oral Language Development, (2) Lan-
guage Learning Strategies, and (3) Personality and Social Factors.
Specific Features of L2 Oral Language Development

Research on L1 oral language development has focused on the acquisi-
tion of specific aspects of the language, such as vocabulary, specific gram-
matical forms, or pragmatic patterns. Our search of the L2 oral language
development literature revealed surprisingly few studies of this type. Thus,
we have a very limited understanding of specific aspects of L2 oral lan-
guage development and, thus, little empirical basis for planning educa-
tional interventions that would promote language development in specific
ways. In light of evidence reviewed in Chapter 3 that specific aspects of
oral language proficiency are linked to literacy and academic development,
this gap in our knowledge is of concern. Notwithstanding the overall lack
P1: PJU
16 William M. Saunders and Gisela O’Brien
of research of this sort, two domains of L2 oral language acquisition have

received some attention; namely, question formation and vocabulary. One
of the major themes that runs through these studies concerns the nature
and development of more academic uses of oral language. In an attempt
to investigate that theme further, this subsection concludes with a review
of results drawn from several studies in the corpus that report correlations
between L2 oral and L2 reading.
Findings from research on question formation suggest that the acquisi-
tion of question forms in ELLs is similar to that observed among monolin-
gual English-speaking children. In addition, more proficient ELLs demon-
strate a wider repertoire of question forms than less proficient ELLs, but
even less proficient ELLs demonstrate some command over English ques-
tion forms and show considerable growth over relatively short periods of
time (six months to a year). Support for this trend comes from research by
Lindholm (1987) and Rodriguez-Brown (1987).
Lindholm (1987) carried out longitudinal and cross-sectional analyses of
natural language samples from young fluent and limited English proficient
(LEP) Spanish-speakers. She found (a) increasing sophistication in ELLs’
questions over a one-year period, (b) question use even among students
with limited oral language proficiency, and (c) significant differences in the
kinds of questions used by limited and fluent English proficient students,
indicating that, with greater proficiency, ELLs acquire increasing command
over more sophisticated question types.
Rodriguez-Brown (1987) found essentially the same pattern of results
as Lindholm among Grade 3 ELLs from bilingual homes who were par-
ticipating in a two-way immersion program. While some students were
fluent-Spanish but LEP and others were fluent-English but limited-Spanish
proficient, the kinds of questions used by these two groups were sim-
ilar in their stronger and weaker languages. These results suggest that
question forms might develop similarly for L2 learners (i.e., regardless
of whether their L1 is Spanish and their L2 is English, or vice versa), at
least for languages that belong to the Indo-European family. Additionally,
students showed equal facility with almost all question types but used
a wider variety of question types in their stronger language. Finally, stu-
dents with low levels of proficiency could transact requests for information
and yes/no questions, suggesting that question use emerges early in L2
development.
Studies on vocabulary development indicate that ELLs demonstrate
greater capacity to define words as they become more proficient. Initially,
ELLs are more likely to define words through simple associations, termed
informal definitions (e.g., cat: “My aunt has one and it’s all furry and has a
long tail.”), and then at higher levels of proficiency through explication,
termed formal definitions (e.g., cat: “A cat is a domesticated mammal which
is related to the lion.”). This line of research, initiated by Snow et al. (1987;
P1: PJU
Oral Language 17
the source for the examples listed previously) is particularly relevant

because it is one, if not the only, attempt in this corpus to operational-
ize and examine empirically the nature of oral language use for academic
purposes. The protocol used by Snow and her colleagues involves asking
students in a one-on-one administration what relatively common words
mean. Their definitions are coded as informal or formal and are rated for
quality. The most effective responses – high quality formal definitions – do
not presume shared knowledge with the interlocutor, do not necessitate
or attempt to elicit interactive support from the interlocutor, and involve
more sophisticated vocabulary and syntax. All three of these elements are,
according to Snow and her colleagues, hallmarks of language that is appro-
priate for interpersonally decontextualized, academic usage.
Snow et al. (1987) found that among middle-class 2nd through 5th
graders, some of whom were ELLs and others native English speakers
learning French, there was a significant correlation between L2 proficiency
and the quality of students’ formal definitions. Moreover, the strength of
the correlation increased over grades: r = 0.16, 0.45, and 0.50 in Grades 2,
4, and 5, respectively. Finally, they found that ELLs with high levels of L2
proficiency scored as well as native English speakers.
Carlisle et al. (1999) corroborated most of Snow et al.’s findings in a
study of low SES Spanish-speaking children in Grades 1, 2, and 3. Carlisle
et al. also helped ground empirically the distinction between formal and
informal definitions. They found a stronger relationship between informal
definitions and receptive vocabulary in both Spanish and English (r =
0.67 in English and r = 0.79 in Spanish) than between formal definitions
and receptive vocabulary (r = 0.36 in English and r = 0.43 in Spanish).
These data fit well with Snow et al.’s original formulation. Both receptive
vocabulary (correctly associating words with picture stimuli) and informal
definitions measure less formal aspects of oral language proficiency. In
contrast, formal definitions measure more formal and academic aspects
of language proficiency. That the pairs of correlations vary substantially
(almost 2 to 1: 0.67 and 0.79; 0.36 and 0.43) suggests that being familiar with
words (including receptive vocabulary and informal definitions) is one
aspect of language proficiency, whereas explicating their meaning (formal
definitions) is qualitatively different.
Several studies report correlations between L2 oral proficiency and L2
reading achievement. Table 2.1 organizes all correlations by study and
grade level. Among the twenty-eight correlations, significant relationships
were more evident among oral measures that are linked to more academic
aspects of language proficiency: specifically, vocabulary (Saville-Troike,
1984), formal definitions (Carlisle et al., 1999; Snow et al., 1987), and story-
retell content (Goldstein, Harris, and Klein, 1993). Goldstein et al.’s story-
retell correlations provide a good illustration. A significant correlation
emerged with their measure of the quality of the content of the retells, as
P1: PJU
table 2.1. Correlations Between L2 Oral and L2 Reading Measures

0521859751c02
18
Grade Levels
Studies, n and Oral English English Reading
Mixed Grade Levels Measure Measure Mixed 1st 2nd 3rd 4th 5th 6th ≥ 7th
• Saville-Troike (1984) Interview: Vocabulary CTBS 0.63a
n = 19; 2nd–5th Interview: Verbosity 0.40
Interview: Syntax 0.29
NSST 0.29
CB989B/Genesee
BSM 0.26
FLS 0.14
Interview: T-Unit Length 0.11
Interview: Grammar 0.03
• Carlisle et al. (1999) PPVT CAT 0.42a
n = 57; 1st–3rd Informal Definitions 0.43a
0 521 85975 1
Formal Definitions 0.33a

• Snow et al. (1987) Formal Definitions CAT 0.16a 0.45a 0.50a
n = 136
• Ulibarri et al. (1981) LAS-O CAT, CTBS, SAT 0.48b 0.29b 0.31b
n = 165–207 BSM 0.26b 0.25b 0.16b
BINL 0.11b 0.15b 0.27b
• Royer and Carlo SVT: Listening SVT: Reading 0.23 0.17
(1991),
n = 29
• Goldstein et al. Story Retell: Content PIAT 0.40a
(1993)
n = 31; 7th–9th SE Story Retell: Language 0.12
• Garcia-Vázquez WLPB ITBS 0.70a
et al. (1997)
n = 100; 6th–12th
November 12, 2005
Notes: a Statistically significant (p < 0.05); b Ulibarri et al. did not report significance levels (zero-order correlations within a multiple regression analysis).
Abbreviations: BINL: Basic Inventory of Natural Language; BSM: Bilingual Syntax Measure; CAT: California Achievement Test; CTBS: California Test of
Basic Skills; FLS: Functional Language Survey; ITBS: Iowa Test of Basic Skills; NSST: Northwest Syntax Screening Test; LAS-O: Oral portion of Language
15:0
Assessment Scales; PIAT: Peabody Individual Achievement Test; PPVT: Peabody Picture Vocabulary Test; SAT: Stanford Achievement Test; SVT: Sentence
Verification Technique; WLPB: Woodcock Language Proficiency Battery. Goldstein et al. (1993) includes Special Education students (SE).
P1: PJU
Oral Language 19
measured by the inclusion of details about the plot, setting, and characters’
intentions, for example (r = 0.40), but not with their measure of the quality
of language used to retell the stories, as measured, for example, by the
use of complete sentences and correct syntax (r = 0.12). This same finding
holds for results elicited using language proficiency test batteries (Garcia-
Vázquez et al., 1997; Ulibarri, Spencer, and Rivas, 1981). More specifically,
broader, more academically oriented batteries like the Woodcock Language
Proficiency Battery (WLPB; r = 0.70) and the Language Assessment Scales-
Oral component (LAS-O; r = 0.29 to 0.48) correlated more strongly with
reading outcomes than assessments like the Bilingual Syntax Measure and
the Basic Inventory of Natural Language (r = 0.11 to 0.27) that have a
narrower, less academic focus.
In a related vein, there is some evidence that the relationship between
reading achievement and measures of English oral proficiency that have an
academic focus becomes stronger in advancing grades, arguably because
both are similarly influenced by schooling and both are indicative of aca-
demic success. More specifically, Snow et al. (1987) found significant,
increasingly large correlations between reading achievement and qual-
ity of formal definitions across Grades 2, 4, and 5: r = 0.16, 0.45, and
0.50, respectively. Garcia-Vázquez et al. (1997) found a correlation of 0.70
between WLPB and reading achievement for bilingual Hispanic students in
Grades 6 to 12, the oldest sample represented and the strongest correlation
reported in Table 2.1.
The relationship between L2 oral proficiency that is linked to academic
uses and academic achievement deserves further research attention, par-
ticularly in the higher grades. As discussed later in this chapter, there is vir-
tually no U.S. research on how classroom instruction might best promote
more academic aspects of oral language development, and there is very
little research on oral language proficiency beyond the elementary grades.
Language Learning Strategies

An added dimension of L2 acquisition is the use of strategies for acquir-
ing language. The use of explicit strategies often characterizes L2 acqui-
sition because ELLs are typically older and more mature than L1 learn-
ers, and they already have competence in an L1. Thus, L2 acquisition
does not call on exclusively implicit processes but can also entail con-
scious or explicit strategies. In this regard, Chesterfield and Chesterfield
(1985a) provide a comprehensive study of the strategies used by pre-
school and early elementary age Spanish-speaking ELLs from Mexican-
American families. The study incorporated strategies that appear to be
directly linked to the acquisition of the target language (language-learning
strategies), as well as strategies that serve L2 acquisition indirectly through
communication with target language speakers (communicative strategies).
P1: PJU
Chesterfield and Chesterfield found a hierarchical relationship among

twelve communicative and language-learning strategies, in ascending
order: repetition, memorization, formulaic expressions, verbal attention
getters, answer in unison, talk to self, elaboration, anticipatory answer,
monitoring, appeal for assistance, request for clarification, and role play.
Strategies at the lower end of the scale (e.g., repetition and memorization)
do not necessarily elicit further interaction and tend to be more receptive.
Those at the middle range of the scale (e.g., verbal attention getters, elab-
oration, and anticipatory answers) serve to initiate and maintain interac-
tion with interlocutors. The higher end of the scale includes strategies that
involve a heightened awareness of language and communication: monitor-
ing, appeal for assistance, and request for clarification. It follows from the
results of this implicational scale analysis that regular use of any particular
strategy implies use of all other strategies lower in the hierarchy.
Chesterfield and Chesterfield (1985a) also found that (a) a substantial
proportion of ELLs’ interactions involved language-learning strategies;
(b) strategies emerged over time in the same relative order, although at
different rates and times, for ELLs who began pre-school with more and
less English proficiency; and (c) students’ repertoires of language-learning
strategies developed along with increasing English proficiency. Overall,
this study contributes to our understanding of language learning by estab-
lishing a relationship among learner strategies that had been studied sepa-
rately in prior research (i.e., communicative and language-learning strate-
gies). It also documents the use of strategies among young ELLs and
thereby confirms some degree of intentionality in their language learning.
One limitation to the Chesterfield and Chesterfield research, and studies
that preceded it, however, is that data were collected almost exclusively
through recordings or observations of students’ interactions with others.
Thus, the strategies that ELLs might use when they are not interacting
with others were not examined. Indeed, Saville-Troike (1988) found that
ELLs use language-learning strategies even during the silent period, when
many ELLs (six of nine in her study) engage in few, if any, interactions
with other ELLs or English speakers. Saville-Troike’s findings with respect
to the use of language-learning strategies during the silent period help
explain at least one mechanism by which some ELLs successfully acquire
English, in some cases at rates and with outcomes that surpass those of
more social and communicative ELLs, despite lengthy periods of little or
no interpersonal communication with English speakers.
We conclude this subsection with discussion of two studies that focused
on the use of learning strategies by high school ELLs (O’Malley et al.,
1985a, provide a summary of both studies; see O’Malley et al., 1985b,
for a more detailed explanation of the first of the two studies). The first
study identified through the use of small group interviews the strategies
students used in different discrete and integrated listening and speaking
P1: PJU
Oral Language 21
tasks; discrete tasks included practicing pronouncing words and learning

vocabulary words, for example; and the integrated tasks included listening
to a lecture and preparing and making an oral presentation. Following pre-
vious L1 research, strategies were classified as metacognitive (e.g., selective
attention), cognitive (e.g., note-taking), and socioaffective (e.g., collaborat-
ing with others). O’Malley et al. found that students reported using cog-
nitive strategies more often than metacognitive strategies, and they were
more likely to report using strategies with less demanding discrete tasks
than with more demanding integrative tasks.
Based on these findings, O’Malley et al. conducted a second study
designed to test the effects of an eight-day (50 minutes per day) interven-
tion designed to train students to use metacognitive and cognitive strate-
gies in the context of integrative tasks: listening to lectures and making oral
presentations. Students were randomly assigned to one of three groups –
metacognitive + cognitive strategies, cognitive strategies only, and con-
trol. Students were pre- and post-tested on both listening to lectures and
making oral presentations. Analysis of covariance (controlling for pretest
variation among the groups) found no significant differences on the listen-
ing task but significant differences favoring the metacognitive + cognitive
strategies group on the speaking task. In fact, O’Malley et al.’s analyses of
the strengths and weaknesses of their training modules suggest that with
refined curricula and delivery, listening outcomes could be improved.
The O’Malley et al. studies are unique in at least two respects. First, like
other studies reviewed in this subsection, their first study demonstrates
that ELLs use explicit and conscious strategies to acquire language and
to accomplish listening and speaking tasks. Second, their second study
demonstrates that ELLs, at least teenage ELLs, might benefit from instruc-
tion and training in the use of strategies. The first study determined that
students were less likely to apply strategies, particularly meta-cognitive
strategies, to more challenging integrative tasks, and the second study
demonstrated that this weakness or need was amenable to instruction and
training. It is difficult to generalize based on the results of one study, but
the evidence recommends further research. We note this particularly in
light of the fact that this was the only study we located that systematically
examined the effects of instruction and training on oral language outcomes.
We include discussion of O’Malley et al. in this section because of its rela-
tionship to language learning strategies. With its focus on instruction and
training, it could also have been discussed in the School Factors section.
Personality and Social Factors

Research on personality and social factors has been motivated by interest
in the individual differences that account for variation in L2 proficiency.
Arguably, evidence for such influences could be useful in educational
P1: PJU
settings to the extent that teachers could modify instruction to better match
individual learners’ learning styles or personal characteristics. Researchers
often postulate that children who are disposed to social interaction might
possess language-learning advantages on the assumption that such chil-
dren seek out more interactions with fluent English speakers and will
thereby engender more English input, interactive experiences and, conse-
quentially, language acquisition opportunities. Wong-Fillmore (1976, cited
in Strong, 1983) added a sociocultural dimension to this hypothesis. Among
the Spanish-speaking ELLs she studied, the strongest language learners
were more willing to engage with L2 speakers in order to gain access and
acceptance from that group.
Research by Strong (1983) has expanded our understanding of the
complexities of the relationship posited by Wong-Fillmore. More specif-
ically, Strong analyzed the “natural communicative language” (NCL) of
thirteen Spanish-speaking ELLs in one bilingual kindergarten class com-
prised of ELLs, fluent English proficient students (FEPs), and monolin-
gual English speakers. NCL was elicited during interviews and observed
during play activities. Audio recordings of these activities were tran-
scribed and analyzed for sentence structure, vocabulary, and pronuncia-
tion. Strong found that NCL measures were unrelated to some social styles
but strongly related to others: talkativeness (initiations toward others in
Spanish), responsiveness (responses to others in Spanish), and gregarious-
ness (number of interlocutors in Spanish and/or English). He also found
that the frequency with which ELLs interacted with native English speakers
was associated with some aspects of language proficiency (viz., vocabu-
lary) but not others (viz., sentence structure and pronunciation). Drawing
on the strong and significant correlations between NCLs and talkativeness
and responsiveness (r = 0.65−0.82), Strong argued that the most successful
language learners maintain interaction more effectively than less success-
ful language learners because they are equally capable of initiating interac-
tion and responding to others’ initiations. Strong argued further that mere
exposure to English speakers is probably not as important as the nature of
the interactions that ensue between ELLs and native English speakers.
In a separate study, Strong (1984) examined another social/personality
factor – namely, integrative motivation – defined as the willingness of the
learner to associate with members of the target language group. Strong
found evidence of a significant, positive relationship between integrative
motivation (indexed by number of nominations of English-speaker play-
mates) and language proficiency among ELLs who began school with rel-
atively high levels of English. However, he found no evidence of such a
relationship among ELLs who began the year with low levels of English
proficiency. Strong proposed that associations with members of the tar-
get language group might result from high levels of proficiency in that
language, rather than the other way around. In fact, when a number of
P1: PJU
Oral Language 23
playmate nominations was plotted across the year, as students’ English

developed, there was an increasing tendency to nominate monolingual
English speakers. These findings suggest that ELLs may need some mini-
mum level of English proficiency before they are likely to begin associating
with monolingual English speakers.
school factors
Perhaps no topic speaks more directly to the education of ELLs than school-
ing. One might have expected to find a fairly large body of research that
examined the effects of different types of programs and instructional mod-
els on ELLs’ oral language development. In fact, no such body of research
exists. For the most part, research on the effects of programs and instruc-
tional models ignores oral language outcomes in favor of literacy out-
comes (see Chapter 4) or academic achievement outcomes (see Chapter 5).
Notwithstanding this significant gap in the literature, studies were identi-
fied that speak to issues related to school and classroom learning contexts
(see Table A.2.2 for summaries of these studies). More specifically, in this
section we review research on two topics: (1) Rates of Oral Language Pro-
ficiency Development, and (2) Language Use and L2 Oral Development.
Rates of Oral Language Proficiency Development

The rates at which ELLs achieve oral language proficiency is of consid-
erable interest at least in part because of the long-standing policy debate
about how long ELLs should receive federally funded services. Estimates
of proficiency attainment typically focus on literacy (see Collier, 1987).
With one exception (Hakuta, Butler, and Witt, 2000), no U.S. study pub-
lished within the last twenty years has explicitly addressed the rates of
oral English language proficiency attainment. However, our search iden-
tified a small number of studies that report longitudinal or cross-sectional
oral language outcomes, providing the opportunity to begin to look at
rates of development of oral proficiency over years of instruction and
schooling.
Table 2.2 displays results reported in six different studies. Studies are
organized by their research designs: one-year and multiple grades, longi-
tudinal, quasilongitudinal, and cross-sectional. For each study, Table 2.2
lists the program students participated in (see table notes for explanation
of abbreviations), the oral language assessment instrument, sample or sub-
samples of students, language of testing, and number of students. Mean
scores are arranged by grade level, from Kindergarten to Grade 5. With one
exception (Hakuta et al., 2000, Sample B), all results are reported in terms
of mean proficiency levels based on a five-point scale (see table notes for
the conversions we performed on some datasets). In general, Level 5 is
P1: PJU
table 2.2. Results of Longitudinal and Cross-Sectional Studies of Oral Language Outcomes
0521859751c02
24
Oral Language Mean
Results by Grade
Sample or Test Overall Per
Study Design Program Instrument Subsamples Lang N K 1 2 3 4 5 Gain Year
Weslander & 1-year multiple ESL BSM ELLs Eng 577 0.43 0.43
Stephany grades Levels 1–5
(1983)
Medina & Longitudinal DBE LAS-O Span-spkrs Eng 298 1.86 3.70 1.84 0.61
CB989B/Genesee
Escamilla DBE Levels 1–5 Limited Span 187 1.45 2.42 0.97 0.32
(1992) Span-spkrs
Howard et al. Longitudinal TWI Proj. dev. Span-spkrs Eng 125 4.35 4.86 0.51 0.26
(2003) TWI Levels 1–5 Eng-spkrs Span 115 3.58 4.14 0.56 0.28
Thomas & Quasi- TWI 50/50 SOLOM Span-spkrs Eng 238 2.72 3.28 3.44 3.49 4.02 1.30 0.33
Collier (2002) longitudinal TWI 50/50 Levels 1–5 Eng-spkrs Span 279 1.57 1.61 2.06 2.73 2.87 1.30 0.33
0 521 85975 1
Lindholm- Cross-sectional TWI 90–10 SOLOM Span-spkrs Eng 1,816 2.82 3.32 3.70 3.74 4.28 4.82 2.00 0.40
Leary (2001) TWI 90–10 Levels 1–5 Eng-spkrs Span 1,008 2.88 3.50 3.80 4.04 4.40 4.80 1.92 0.38
Sample H
Lindholm- Cross-sectional TWI 90–10 SOLOM Span-spkrs Eng 1,122 2.64 3.12 3.54 3.88 3.96 4.48 1.70 0.34
Leary (2001) TWI 90–10 Levels 1–5 Eng-spkrs Span 707 2.52 3.36 3.72 4.04 3.96 4.14 1.62 0.32
Sample L
Hakuta et al. Cross-sectional All-Eng IPT ELLs Eng 1,872 1.75 3.40 4.35 4.60 4.80 3.05 0.76
(2000) Levels 1–5
Sample A
Hakuta et al. Cross-sectional Varied WLPB-R ELLs Eng 122 4.4 5.8 7.5 3.1 0.78
(2000) Age Equiv. Eng-spkr Eng 7.0 9.0 11.0 4.0 1.0
Sample B norms
Notes: Program designations are ESL: English as a Second Language; DBE: Developmental Bilingual Education program; TWI: Two-way Immersion program (% Sp
instruction in K/% English instruction in K); All-English: no primary language instruction. The following conversions were calculated by chapter authors: BSM results
November 12, 2005
were converted from a 6- to a 5-point scale; SOLOM raw means were converted to mean levels; IPT means, originally expressed on a scale from 0 to 1.0, were converted
to a 5-level scale (IPT is based on a 6-level scale). Thomas and Collier’s results are based on years in program (1–5, quasilongitudinal); K-6 elementary school, Northwest,
U.S. Lindholm’s Sample H includes students from schools with high ethnic density and high SES need, and Sample L includes students from schools with low ethnic
15:0
density and SES need. Hakuta el al.’s Sample B includes students from both bilingual and all-English programs; the study found no significant differences between
the two groups and collapsed them into one sample.
P1: PJU
Oral Language 25
interpreted as native-like proficiency, and Level 4 is viewed as sufficient for

participation in mainstream English instruction – that is, proficient but not
yet native-like. Expressing results on a common scale allows us to compare
results across studies, samples, and subsamples. Overall and per year gains
were calculated and are displayed in the far left-hand columns of the table.
Results for Sample B from Hakuta et al. (2000) are explained at the end of
this subsection.
Three studies (Howard, Christian, and Genesee, 2003; Lindholm-Leary,
2001; Thomas and Collier, 2002) include both Spanish-speaking ELLs and
native or fluent English speakers involved in two-way or dual language
immersion programs, wherein both ELLs and native-English speaking stu-
dents learn a second language. Each study reported L2 oral language out-
comes for all students: English for Spanish-speaking ELLs and Spanish for
native or fluent English speakers. Results for these two subsamples allow
for comparisons across different L2s: Spanish and English.
The data reported in these studies vary in terms of the number of
data points, design, sample characteristics, programs under investigation,
and the instruments used to assess oral proficiency, including criterion-
referenced assessments (Bilingual Syntax Measure – BSM, Language
Assessment Scales-Oral – LAS-O, Howard et al.’s experimenter-developed
instrument, and the Idea Oral Language Proficiency Test – IPT), a teacher
rating scale (Student Oral Language Observation Matrix – SOLOM), and a
norm-referenced test (Woodcock Language Proficiency Battery-Revised –
WLPB-R). Thus, caution must be exercised in interpreting the following
patterns. Nevertheless, the results compiled in Table 2.2 represent, to the
best of our knowledge, the first attempt to synthesize results across stud-
ies and examine patterns in the rates at which ELLs attain oral language
proficiency.
A number of noteworthy trends emerge from analysis of these data.
First, it seems that ELLs, on average, require several years to develop oral
English proficiency. More specifically, the results in Table 2.2 show that
means of 4.00 or higher (generally proficient but not yet native-like) do
not emerge before Grade 3 and do not appear consistently across studies
until Grade 5. Even in the dataset that includes students from all-English
programs (Hakuta et al., Sample A), where presumably students receive
maximum exposure to English, means of 4.00 or better do not appear
until Grade 3. Moreover, none of the datasets include means that begin to
approach 5.00 (native-like) until Grade 5. This holds regardless of whether
students participated in bilingual (Lindholm-Leary, Howard et al., Thomas
and Collier) or all-English programs (Hakuta et al., Sample A). Second, and
in a related vein, ELLs, on average, tend to make more rapid progress from
lower to middle levels of proficiency (i.e., from Levels 1 through 3; see
especially Medina and Escamilla, and Hakuta et al., Sample A) and slower
progress as they move beyond Level 3. By the end of Grade 3, means are
P1: PJU
typically at or around 4.00, but it takes until Grade 5 before means begin
approaching 5.00 (Hakuta et al., Sample A; Lindholm-Leary, Sample H;
Howard et al.).
Third, despite varied measures, samples, programs, and even lan-
guages, rates of L2 oral language progress appear to be strikingly consis-
tent. In most cases, datasets with K or Grade 1 means below 2.00 show larger
per-year gains (Medina and Escamilla: 0.61; Hakuta et al., Sample A: 0.76),
and datasets with K and Grade 1 means above 2.00 show smaller per-year
gains (Thomas and Collier: 0.33; Lindholm-Leary, Sample H: 0.40). How-
ever, notwithstanding these differences, the range in rates of development
is generally small. Of greatest interest in terms of consistency are the results
from two-way bilingual studies that show virtually identical average per-
year gains among Spanish-speakers tested in English and English-speakers
tested in Spanish: respectively, 0.26 and 0.28 per year (Howard et al., 2003),
0.33 and 0.33 (Thomas and Collier, 2002), 0.40 and 0.38 (Lindholm-Leary,
Sample H), 0.34 and 0.32 (Lindholm-Leary, Sample L). Even the average
per-year gain of an intensive ESL program (Weslander and Stephany, 1983)
is within approximately the same range as the other programs: 0.43. The
same holds for the all-English program (Hakuta et al., Sample A), when the
large increase from Grade 1 to 2 (1.75 to 3.40) is isolated and the per-year
gain is recalculated based on means from Grades 2 through 5: 0.47 per year.
Identifying and analyzing reliable estimates of per-year gains no doubt
require more systematic sampling of programs and grade levels, more con-
sistent forms of measurement, and more discerning statistical analyses.
Notwithstanding these cautions, at least two hypotheses might account
for the apparent consistency in per-year gains evident in the existing data:
(a) on average, L2 oral language development proceeds at a fairly constant
rate independent of program; or (b) on average, school contexts, indepen-
dent of program, exert a fairly constant or homogenizing effect on oral
language development.
Hakuta et al. (2000; Sample B), the final dataset listed in Table 2.2, makes
a unique contribution to this discussion of rates of L2 oral development.
Hakuta et al. report cross-sectional proficiency data (age equivalents) for
randomly selected Grades 1, 3, and 5 ELLs based on the WLPB-R. These
data differ significantly from the other data reported in Table 2.2 for two
reasons. First, WLPB-R is generally viewed as more academically oriented
than most available oral proficiency instruments. Second, WLPB-R is a
norm-referenced assessment with norms based on the performance of
native English-speaking children. As Hakuta et al. explain, “[WLPB-R]
was selected because it was felt to be the best measure available to indi-
cate the student’s academic competitiveness with English-speaking peers”
(p. 6). Indeed, Hakuta et al.’s Sample B results stand in sharp contrast to
other results in Table 2.2 that show ELLs at or close to proficient by the
end of Grade 5. Despite gains from 1st to 3rd to 5th Grade, Hakuta et al.’s
results show ELLs performing substantially below English-speaker norms
P1: PJU
Oral Language 27
at all grades, and the gap between ELLs and English-speakers actually
widens from 1st to 5th Grade. By Grade 5, ELLs are performing 3.5 years
below native-English speaker norms. We cannot rule out the possibility
that the below-norm results are simply unique to Hakuta et al.’s sample.
In the meantime, the question remains: Do the criterion-referenced mea-
sures and teacher rating scales commonly used in research and at school
sites underestimate or establish a low ceiling for proficiency, one that falls
far below native English-speaker norms?
Language Use and L2 Oral Development

Several studies in the corpus examined ELLs’ language use in the class-
room. Based on the assumption that language use contributes to language
development, researchers have investigated ELLs’ language choices, the
nature and outcomes of peer interactions involving ELLs and fluent English
speakers, and the relationship between L2 use and L2 proficiency. All of
the studies to be discussed focused on language use during interactive
classroom activities, including paired activities, cooperative groups, and
independent work time when students are allowed to converse and assist
one another. During these activities, ELLs were free to choose to use either
L1 or L2.
In general, the evidence suggests that ELLs’ language choices tend to
align with the dominant language of instruction. Chesterfield, Chesterfield,
Hayes-Latimer, and Chávez (1983) investigated the language choices of
Spanish-speaking ELLs in bilingual pre-school classes. In classes where
teachers tended to use more English throughout instruction, ELLs tended
to use more English with their peers. In classes where teachers tended
to use more Spanish, ELLs tended to use more Spanish. Chesterfield and
Chesterfield (1985b) also report language use data for Grade 1 Mexican-
American ELLs, half of whom were enrolled in “English” classes and half
of whom were enrolled in Spanish bilingual classes. In the English classes,
ELLs used English during peer interactions a majority of the time. ELLs in
the bilingual classes used Spanish a majority of the time. Among Grade 2
ELLs in Spanish bilingual programs where at least a majority of instruction
was delivered in Spanish, both Milk (1982) and Malave (1989) found that
ELLs were more likely to use Spanish during peer interactions; in fact,
Malave found students using Spanish over English by a ratio of 6 to 1.
Finally, among Grade 4 ELLs who had participated in Spanish bilingual
classrooms through Grade 3 and were then placed in an “English-only”
class, Pease-Alvarez and Winsler (1994) found a substantial increase from
the beginning to the end of the year in students’ use of English in their
classroom interactions (53 to 83 percent).
While the studies reviewed thus far focused on language choice during
classroom interactions, another topic represented in the corpus examines
explicit attempts to cultivate interaction between ELLs and native or fluent
P1: PJU
English speakers. Most programs for ELLs incorporate some provision for
the integration or mixing of ELLs and native or fluent English speakers
(see Genesee, 1999). The assumption is that such integration, aside from its
potential social benefits, provides ELLs with worthwhile language learning
opportunities. The corpus of peer-interaction studies, however, suggests
that creating such opportunities and producing positive oral language out-
comes involve more than simply pairing ELLs with native or fluent English
speakers.
Several studies suggest that pairing native or fluent English speakers
and ELLs, in and of itself, may not yield language-learning opportuni-
ties. For example, in a study of naturally occurring interactions among
kindergartners during play situations, Cathcart-Strong (1986) found that
the response patterns of native English-speaking peers did not depend-
ably provide interactions that would be expected to contribute to ELLs’
language development. She concluded that such interactions might only
come from adult interlocutors or in response to more carefully structured
tasks. In a related vein, Platt and Troudi (1997) describe the case of a Grebo-
speaking girl enrolled in a mainstream English classroom where the teacher
relied almost exclusively on native-English-speaking students to support
ELLs’ classroom participation. In fact, the ELL child’s interaction with
her native-English-speaking partners rarely provided language learning
opportunities, primarily because class assignments were well beyond her
language and knowledge, and her English-speaking peers were at a loss as
to how to assist her. Similarly, Jacob et al. (1996), in a study of cooperative
learning groups comprised of Grade 6 ELLs and native English speakers,
found few instances that served as language learning opportunities for
ELLs. The researchers concluded that interaction in cooperative groups is
heavily influenced by the nature of the tasks and by the students’ interpre-
tations of the tasks. In this class, ELLs and native English speakers tended
to cut short their interactions in order to complete assigned tasks in the
allotted time: “Just write that down. Who cares? Let’s finish up” (Jacob
et al., p. 270).
Other studies (August, 1987; Peck, 1987) confirm the important role of
tasks and also the training required of native English speakers to help
them become language-learning facilitators. However, they also suggest
that there is probably a minimum level of oral proficiency ELLs require in
order for them to benefit from structured paired activities, at least in terms
of verbal participation. August (1987) employed specific tasks to guide
interaction between six- to ten-year-old ELLs and fluent English speakers
and found a significant relationship between the frequency of verbal inter-
actions and L2 proficiency. ELLs with relatively high levels of proficiency in
contrast to ELLs with lower levels of proficiency interacted more frequently
and extensively with fluent English peers. Similarly, Peck (1987) carefully
selected and then trained a Grade 2 native English speaker to teach games
to nine Kindergarten Spanish-speaking ELLs of varying proficiency levels.
P1: PJU
Oral Language 29
Language outputs correlated with proficiency levels such that high- and
middle-level students, in comparison to low proficiency students, spoke
significantly more and displayed more sophisticated vocabulary.
Johnson (1983) tested the effects of a program (Inter-ethnolinguistic
Peer Tutoring, IEPT) that incorporated specific tasks and training designed
to promote more extensive interaction between ELLs and fluent English
speakers (FESs). The study was conducted over a five-week period.
Matched pairs of ELLs and matched pairs of FESs (16 ELLs and 18 FESs;
5–9 years old) were randomly assigned to either the treatment or control
group. Interactions before, during, and after the five-week treatment were
observed and coded. Data were analyzed for group differences over time
and at the end of the treatment period. Results favored the treatment group
but also revealed an interesting pattern that emerged over time. By the end
of the five-week period, ELLs and FESs in the treatment group were inter-
acting significantly more than ELLs and FESs in the control group. How-
ever, the difference emerged over time because verbal interactions between
ELLs and FESs increased slightly for students in the treatment group but
declined considerably for students in the control group. At least in this
case, the treatment seemed to help ELLs and FESs maintain interactions
that – to the extent the control group is representative – typically taper off
over time.
A corollary issue concerning language use is whether increased L2
use results in increased L2 oral language proficiency, on the one hand,
and enhanced academic achievement on the other. Two of the studies
already discussed also analyzed L2 oral outcomes associated with L2 use
(Chesterfield et al., 1983; Johnson, 1983), and another study examined rela-
tionships among L2 use, oral outcomes, and reading achievement (Saville-
Troike, 1984). The findings of these studies are suggestive but by no means
conclusive. Findings from the Chesterfield et al. and Saville-Troike stud-
ies, in particular, help unpack and also qualify the potential effects of
L2 use.
Johnson (1983) administered three oral language proficiency assess-
ments to ELLs prior to and immediately following the five-week IEPT
treatment: Language Assessment Scales (LAS), Peabody Picture Vocabu-
lary Test (PPVT), and the Child-Child Communication Test (CCCT). To
investigate the relationship between the amount of interaction in English
and growth in English proficiency, Johnson calculated partial correlations
(controlling for pre-test) between post-test proficiency scores and mea-
sures of the frequency of ELLs’ interactions with FEPs. None of the partial
correlations, however, was statistically significant. Johnson also compared
pre- to post-test gains of the treatment and control ELLs on the PPVT and
the LAS. A significant difference favoring the treatment group was found
on the PPVT but not on the LAS. Thus, IEPT contributed to vocabulary
growth, but within the five-week period there was no identifiable relation-
ship between individual students’ L2 use and their gains in L2 proficiency.
P1: PJU
In contrast, among pre-school ELLs and over a longer duration of time,

Chesterfield et al. (1983) found significant and positive rank-order correla-
tions between increased L2 use and increased L2 oral proficiency. However,
results suggest that while there may be a positive relationship between
English use and oral gains, ELLs (particularly young ELLs) might bene-
fit differentially from teacher and peer interlocutors depending on their
level of L2 proficiency. Among relatively less proficient students, gains
in L2 proficiency (mean length of utterances) correlated significantly with
increased interactions with the teacher (rho = 0.93). Among relatively more
proficient students, gains in proficiency were significantly correlated with
increased interactions with peers (rho = 0.83).
Saville-Troike (1984) found positive and significant rank-order corre-
lations between English use and oral proficiency among nineteen ELLs
ranging in age from 7 to 12 (Grades 2–6). Students’ parents were partici-
pating in a one-year graduate program for foreign students. The year of
the study represented the first year of L2 exposure and instruction for all
nineteen ELLs. The overall amount of students’ English verbal interactions
(percentage of English interactions sampled across an entire 5.5 hour day)
correlated significantly with their end of year language proficiency rank-
ings on the Northwest Syntax Screening Test (r = 0.69) and with their mean
length of T-units determined through interviews (r = 0.50).
However, while Saville-Troike found that L2 oral language proficiency
was associated with overall L2 use, there was little evidence among her
first-year ELLs of any relationship between use or proficiency and aca-
demic achievement. Specifically, none of the study’s measures of English
language use correlated significantly with standardized measures of
English reading (CTBS): total interactions (r = 0.19), interactions with child
(r = −0.06), and interactions with adults (r = 0.13). With one exception,
moreover, none of the study’s measures of language proficiency correlated
significantly with CTBS reading: NSST (r = 0.29), Functional Language
Survey (r = 0.14), Bilingual Syntax Measure (r = 0.26), interview-based
measures of verbosity (r = 0.40), mean length of T-unit (r = 0.11), and
grammatical accuracy (r = 0.03). Only vocabulary (the number of different
vocabulary items students used during interviews) correlated significantly
with CTBS reading: r = 0.63.
Saville-Troike herself identifies a number of possible reasons for the
failure to find a link between L2 oral language use and achievement.
First, the oral language proficiency instruments in her study may not
have captured the kind of language proficiencies that are implicated by
academic tasks and tests (like CTBS reading). Second, interpersonal com-
munication, despite or perhaps because of its social/contextual nature,
may not be a particularly fruitful means of promoting oral language pro-
ficiency in the service of academic achievement, at least not for students
who are just beginning to learn L2. In many cases, students’ interpersonal
P1: PJU
Oral Language 31
communication with each other and with native English-speaking peers, as

well as with teachers, could be successfully negotiated with gestures and
single words, providing little opportunity for academic language devel-
opment. Third, among the students who most quickly began initiating
and carrying out interactions with peers in English, not one was in the
group that scored highest on the CTBS. To the contrary, the five students
who scored highest on the CTBS at the end of the year used little or no
English at all in their interactions with peers throughout the whole first
half of the year. In fact, three of these five students relied exclusively on
their L1 to clarify and discuss academic assignments during the first half
of the year. Saville-Troike speculates that, at least with this population of
first-year ELLs, access to other L1 speaking peers and adults might make
a stronger contribution to academic achievement than opportunities to
engage in interpersonal communication in English.
non-school factors
The most extensively documented non-school factor related to oral lan-
guage development is language use outside of school, specifically at home
with family and also among peers. The research investigates the relation-
ship among oral English language proficiency, out-of school English lan-
guage use, and L1 maintenance. Although the number of studies reviewed
is small (see Table A.2.3 for a summary of studies), they yield a fairly
straightforward finding – English language use outside of school is posi-
tively associated with ELL’s oral English development. At the same time,
the corpus of studies also begins to unpack the very complex relationship
or tension between learning English on the one hand and maintaining L1
on the other.
Pease-Alvarez (1993) collected self-reports on immigration history and
language use from the parents of fifty-five Grade 3 ELLs of Mexican
descent. Four groups were established ranging from “child and parents
born in Mexico/speak mostly Spanish in the home” (Group 1) to “child
and at least one parent born in United States/speak mostly English in the
home” (Group 4). Oral English proficiency results based on the English
Peabody Picture Vocabulary Test (EPPVT) were closely associated with
immigration history and home language use, with means increasing suc-
cessively from Groups 1 to 4: means = 52, 62, 85, and 90, respectively.
Umbel and Oller (1994) examined samples of thirty-four children at
each of Grades 1, 3, and 6 matched for SES (middle class), parental edu-
cation (high school grad or better), and parental occupation (3 or 4 on
four-point scale, where 3 = sales and clerical workers, and 4 = profession-
als and managers). The parents of all students in the sample spoke both
Spanish and English and introduced their children to both languages at
birth (simultaneous bilinguals). Across the sample, students whose parents
P1: PJU
reported using Spanish and English equally at home averaged about 0.4
standard deviations higher on the EPPVT than students whose parents
reported using more Spanish than English (mean for Spanish-English
equally is 97; for more Spanish than English is 90; sample standard devi-
ation, 0.16). This sample is unusual in that the children are simultaneous
bilinguals who had exposure to both languages in the home since birth.
That is not necessarily typical of the ELL population at large. Nevertheless,
it is noteworthy that an association between English use and proficiency
emerged in such a population.
Hansen (1989) examined the learning rates of 117 Spanish-speaking
ELLs in Grades 2 to 5 in bilingual classrooms over a one-year period using
English auditory vocabulary (Stanford Diagnostic Reading Tests, SDRT) as
an outcome measure. Students were assessed in the fall, spring, and again
the following fall. Using regression analyses, Hansen found that the pro-
portion of English and Spanish used in the home (1 = exclusively English
and 7 = exclusively Spanish) was the best second-order predictor of sub-
sequent English vocabulary achievement, better than language use with
peers or language use in the classroom. Prior English vocabulary scores
accounted for 65 percent and home language use accounted for an addi-
tional 10 percent of the variance in subsequent English vocabulary scores.
While all three of the studies described herein present evidence of a
positive relationship between English use outside school and oral English
proficiency, there are at least three qualifiers that need to be taken into
account. First, although the development of L2 oral proficiency is likely
aided by L2 use outside of school, it is not necessarily impeded by contin-
ued development and use of L1. Second, at higher levels of language and
literacy development, English use outside of school may not be as critical
as English use in school. Finally, L1 and English use in the home are inter-
related, and the nature of the relationship and the impact on children’s L1
and English proficiency are likely mediated by sociocultural factors.
Regarding continued development and use of L1, while Umbel and
Oller (1994) found significant mean differences in English vocabulary
attributable to differences in English-Spanish use at home, they also found
that students with better Spanish vocabularies tended to have strong
English vocabularies. The results of regression analyses indicated that
Spanish vocabulary, followed by grade level, and then home language
use, accounted for, successively, 27, 33, and 36 percent of the variance in
total English vocabulary scores.
Regarding the limits of English use outside school, while Hansen (1989)
found that English use at home contributed to English receptive vocabu-
lary, he found a more complex relationship between various language use
variables and reading comprehension. While the proportion of English
and Spanish used in the home remained a significant predictor in some
analyses, classroom language use and task-oriented peer language use
also proved to be significant and, in some cases, superior predictors of
P1: PJU
Oral Language 33
L2 reading comprehension. Hansen concluded that while home language

use makes a critical contribution to oral language development in general,
language use at school probably plays a more critical role in supporting
higher levels of language and literacy development.
With regard to the possible role of sociocultural factors on L1 use in
the home, we revisit Pease-Alvarez (1993) and Umbel and Oller (1994).
Among working-class families of Mexican descent that varied in terms of
immigration history and Spanish-English use in the home, Pease-Alvarez
found an inverse relationship between English use and proficiency on
the one hand, and Spanish use and proficiency on the other hand. Across
four groups that ranged from “child and parents born in Mexico/speak
mostly Spanish in the home” to “child and at least one parent born in
United States/speak mostly English in the home,” Pease-Alvarez found an
increasing progression in English vocabulary scores (from 52 to 62 to 85 to
90 for Groups 1–4, respectively) and a decreasing progression in Spanish
vocabulary scores (from 71 to 61 to 52 to 46 for Groups 1–4, respectively).
In contrast, among middle-class, bilingual families of Cuban descent
who began introducing their children to Spanish and English at birth,
Umbel and Oller found no relationship between English and Spanish
use and proficiency. Half the sample reported using English and Spanish
equally and the other half reported using more Spanish than English at
home. Children from the former group demonstrated higher levels of
English proficiency, but there was no difference between groups in terms
of Spanish proficiency. Means were identical and virtually at norm (i.e., 96
with a norm of 100) for this sample of simultaneous bilinguals.
The contrasting findings of Pease-Alvarez (1993) and Umbel and Oller
(1994) suggest that the relationship between L1 and English use and its
impact on L1 and English proficiency might be mediated by cultural or
class variables insofar as their respective samples varied along those lines –
working-class families of Mexican descent and middle-class families of
Cuban descent. We located no studies that shed light on the potential cul-
tural differences in this matter. However, we found one study that provides
self-report data indicating that attitudes toward L1 maintenance can vary
across classes within the same culture. Lambert and Taylor (1996) found dif-
ferences between middle- and working-class Cuban-born mothers living
in Miami in their perceptions of the importance of their children learning
English and Spanish. While mothers in both samples reported strong
overall support for learning English and maintaining Spanish, working-
class mothers tended to place greater emphasis on learning English, and
middle-class mothers tended to place greater emphasis on maintaining
Spanish. Working-class mothers tended to associate learning English with
improving one’s economic status; their ratings of their children’s English
fluency correlated positively and significantly with their own perceived
economic status. No such correlation emerged among middle-class moth-
ers. Rather, middle-class mothers tended to associate maintaining Spanish
P1: PJU
with maintaining the family’s cultural identity; their ratings of their chil-
dren’s Spanish fluency correlated positively and significantly with their
own sense of self-respect. No such correlation emerged among working-
class mothers.
The small number of studies reviewed in the section limit the generaliz-
ability of the findings. However, it seems quite clear that the relationship
among L2 development, L1 maintenance, and language use is quite com-
plex. (See also Schecter & Bayley, 1997, for a case study analysis.)
assessment
The assessment of oral English language proficiency is commonplace in
U.S. school districts that serve ELLs. In a survey of ninety-three randomly
selected school districts from the ten states with the highest numbers of
ELLs, Cardoza (1996) found that virtually all districts assessed oral English
proficiency at both program entry and exit. For those students whose par-
ents indicated upon enrollment that a language other than English was spo-
ken in the home, 90 percent of the school districts administered an English
oral language proficiency assessment to determine students’ language clas-
sification and program placement. Eighty-eight percent of the districts also
reported using an English oral language assessment as part of their pro-
cedures for reclassifying students as fluent English proficient (FEP) and
determining whether or not to place students in mainstream English pro-
grams. The vast majority of districts also reported using an English reading
assessment at both entry (72 percent) and exit (93 percent), and most said
they assessed L1 oral language proficiency at entry (64 percent). Only 34
percent of the districts, however, reported assessing L1 oral language at
exit. The most commonly used instruments are the Language Assessment
Scales (LAS), Bilingual Syntax Measure (BSM), and the Language Assess-
ment Battery (LAB). More than 80 percent of the districts reported using
one or more of these three instruments for entry and/or exit assessments.
Other instruments mentioned, though with substantially less frequency,
included the Basic Inventory of Natural Language (BINL) and the Individ-
ualized Developmental English Activities Placement Test (IDEA).
While the assessment of oral language proficiency is commonplace,
research on the validity and reliability of commonly used oral language
assessments is not. Our search identified six studies, most of which focus
on issues related to validity. In this section, we examine validity from the
following perspectives: technical information provided by test developers,
intertest correlations, predictive validity, classifications of students, use of
monolingual test norms, and monolingual versus bilingual assessments.
The section closes with a discussion of one study that illustrates the chal-
lenges involved in developing reliable and valid oral language measures. A
summary of the studies reviewed in the section is presented in Table A.2.4.
P1: PJU
Oral Language 35
One should not assume that the technical qualities of commonly used
oral language assessments have been rigorously or successfully evaluated
and documented. Merino and Spencer (1983) evaluated the technical man-
uals for the English versions of several oral language assessments: LAS,
BSM, LAB, BINL, and the Bahia Bay Area Oral Language Test (BOLT).
Nine items were evaluated for each assessment: Classifications, Criterion-
Validity, Content-Validity, Construct-Validity, Test-retest Reliability, Inter-
scorer Reliability, Internal Consistency, Inter-examiner-Reliability, and,
finally, Norms. The quality of documentation on the tests was evaluated
on a four-point scale: no information provided, information provided
but inadequate, information provided but not thorough, and informa-
tion thorough and adequate. The BINL and LAB provided information
on only three of the items. The BOLT had information on almost all
items, but none were judged satisfactory. The BSM and LAS provided
information on at least six of the nine items, but neither received con-
sistent satisfactory marks: BSM I received satisfactory marks on five of
nine, BSM II on three of nine, LAS I on four of nine, and LAS II on
only one of nine. Based on this review, Merino and Spencer concluded
that schools should use none of the assessments to gauge students’ pro-
ficiency in English. It is possible that the technical information provided
with current editions of these assessments is more complete and adequate
on a wider range of criteria. However, we know of no recent and simi-
larly comprehensive study of these or other currently used oral language
instruments.
The validity of tests is often established by examining their correlations
with similar instruments reputed to assess the same skill or skills. Oral
English language measures typically used to evaluate ELLs do in fact cor-
relate with one another, despite the fact that they often emphasize different
elements of oral language. For example, based on assessments of ELLs in
Grades 1, 3, and 5, Ulibarri et al. (1981) found significant moderate to strong
positive correlations (r = 0.76, 0.74, and 0.58 at Grades 1, 3, and 5, respec-
tively) between the oral subtest scores of LAS, which evaluates vocabu-
lary, syntax, and story retelling, and BSM, which focuses exclusively on
syntax. Based on assessments of Kindergarten and Grade 2 ELLs, Schrank,
Fletcher, and Alvarado (1996) found significant and positive correlations
(r = 0.75 to 0.91) among the oral subtest scores of LAS (and Pre-LAS for K),
the WLPB-R, and the IPT-I. The highest correlations all involved WLPB-R:
0.91 with Pre-LAS among kindergartners, and 0.81 with LAS and IPT-I for
second-graders.
Another way to examine the validity of assessments is to evaluate
the consistency with which they classify students into specific levels of
performance. Significant positive correlations mean that different assess-
ments rank-order students from low to high similarly. However, most oral
language assessments come with recommended cutoff points for different
P1: PJU
levels of proficiency (e.g., non-proficient, limited-proficient, fluent-English

speaking: NES, LES, FES). Thus, when states, districts, schools, and/or
researchers choose a particular oral language instrument, they simulta-
neously choose a set of criteria and cutoff points for classifying ELLs.
Ulibarri et al. (1981) found dramatic differences in the classifications pro-
duced by LAS and BSM. Despite their moderate to strong positive cor-
relations, the two assessment instruments utilized substantially different
cutoff points. Indeed, across more than one thousand ELLs in Grades 1,
3, and 5, LAS and BSM produced the exact same classification (NES, LES,
and FES) for only 32 percent of the students: 51 percent of the sample was
classified as “fluent” by LAS but “limited-proficient” by BSM; 17 percent of
the sample was classified as non-proficient by LAS but limited-proficient
by BSM.
MacSwan, Rolstad, and Glass (2002) conducted an evaluation that
focused specifically on the validity of “non” proficient classifications. When
ELLs are assessed in English upon entry to school, non-proficient classifi-
cations are quite common. However, MacSwan et al. analyzed the Pre-LAS
results of Spanish-speaking Pre-K and Kindergarten ELLs who were classi-
fied as non-proficient in Spanish (n = 6,118 of a total of 38,000; 15 percent).
Among other things, the study found that most of the non-proficient classi-
fications were attributable to students’ difficulty with the last two portions
of the test (sentence completion and story retell). Sixty-seven percent of
the students classified as non-proficient scored 80 percent or higher on
parts 1–3, and 51 percent scored 80 percent or higher on parts 1–4 (acting
out commands, naming items in pictures, identifying pictures that corre-
spond to uttered phrases, and repeating sentences). Space does not allow
for a full explication of their discussion of weighting, scoring, and adminis-
tration issues that also contributed to non-proficient classifications. How-
ever, MacSwan et al. provide sufficient evidence to recommend cautious
interpretation of non-proficient classifications. That this study focuses on
misclassifications in L1 (possibly 50–67 percent false negatives) brings the
problem of misclassification into sharp relief. Undoubtedly, similar mis-
classifications occur with English assessments as well.
Predictive validity refers to the accuracy with which a test predicts per-
formance on other target measures. For example, do oral language mea-
sures predict success in mainstream English classrooms? Results from
both Schrank et al. (1996) and Ulibarri et al. (1981) suggest that more
academically oriented instruments and instruments that sample a broad
range of oral language skills produce stronger correlations with estimates
of mainstream success than less academic instruments with a narrower
focus. Schrank et al. (1996) used teachers’ ratings of students’ academic
language proficiency as an estimate of success in mainstream classrooms
and correlated those ratings with students’ scores on WLPB-R, LAS (and
Pre-LAS for K), and IPT. WLPB-R, typically viewed as a more academically
oriented instrument, yielded the highest correlations with teachers’ ratings
P1: PJU
Oral Language 37
(r = 0.80), and IPT-I, typically viewed as less academically oriented, yielded

the lowest (r = 0.68).
Ulibarri et al. (1981) conducted two separate analyses to gauge the pre-
dictive power of LAS and BSM. The former provides a broader assessment
of oral proficiency, focusing on vocabulary, syntax, and story retelling,
while the latter is narrower, focusing on syntax only. Based on classifi-
cations produced by each measure, Ulibarri et al. conducted analyses of
variance with standardized measures of English reading and math for each
of Grades 1, 3, and 5. LAS classifications produced significant results in the
expected direction for both reading and math at each of the three grade
levels: NES, LES, FES. BSM produced significant differences only between
the FES and LES groups and only in four of the six analyses. Ulibarri et al.
then conducted stepwise regression analyses designed to predict reading
and math achievement at each grade level, using as predictors teachers’
ratings of student achievement and preparedness for mainstream and an
oral language measure (either LAS or BSM). For every regression analysis
(Grades 1, 3, and 5 for each of reading and math), teachers’ ratings proved
to be a better single predictor than either LAS or BSM. However, along
with teachers’ ratings, LAS results added significantly to the prediction
in four of six analyses (overall R-squares ranged from 0.34 to 0.56; LAS
contributed from 0.03 to 0.08 to the overall R-square). BSM did not add
anything significant over and above teacher ratings. Taken together, both
studies suggest that mainstream placement decisions are likely best served
by a combination of informed teacher judgments and oral measures that
are academically oriented and broad in scope.
Some oral language proficiency assessments publish norms, which
allow test users to compare students’ results to that of the norming group.
At least one study suggests that comparing the oral language results of
ELLs to norms derived from monolingual samples can be problematic.
Fernandez et al. (1992) report two important findings based on their anal-
ysis of Spanish and English receptive vocabulary results (Peabody Picture
Vocabulary Test-Revised, PPVT-R, and the Test de Vocabulario en Imágenes
Peabody-Adaptación Hispanoamericana, TVIP-H). Their sample consisted
of 396 Spanish-speaking preschoolers from Miami, two thirds of whom
were of Cuban descent. First, the Cuban American students in the study
sample responded differently to the list of Spanish words on the TVIP-H
than the Mexican and Puerto Rican children who comprised the norming
sample. Some words that were difficult for the study sample were easy
for the norming sample; other words that were easy for the study sample
were difficult for the norming sample. Although both groups were native
Spanish speakers, the Spanish lexicon acquired by students in Mexico and
Puerto Rico and that acquired by the students in Miami varied, making
comparisons between the study sample and the norm group problem-
atic. The implication here is that practitioners or researchers would not
want to interpret norm-referenced results, either for tests of L1 or L2 oral
P1: PJU
language without some knowledge about the composition of the norm

group and its relative comparability to the students being tested.
Second, Fernández et al. also postulate that independent L1 and
L2 receptive vocabulary assessments of young bilingual students (here,
Spanish-English bilinguals) likely underestimate total vocabulary in cases
where a child knows particular words in one language but not the other.
Citing data from another study (Umbel et al., 1992), Fernandez and her col-
leagues estimate that approximately 75 percent of a young bilingual child’s
vocabulary might be constituted by pairs (e.g., English and Spanish ver-
sions of the same concept), and as much as 25 percent in each of L1 and L2
might be comprised of singlets (concepts lexicalized in one language but
not the other). L1 assessments fail to capture the concepts lexicalized only
in L2, and L2 assessments fail to capture the concepts lexicalized only in L1.
The authors argue that a more comprehensive assessment of the linguistic
development of bilingual children should account for vocabulary across
both languages and employ norms based on a bilingual population, not
monolingual samples from each language.
A final topic related to oral language assessments concerns how oral
language is measured. All of the assessments discussed thus far consist of
tasks designed to elicit and evaluate samples of language use under on-
demand conditions. In general, this is no different than most other assess-
ments used in schools to measure reading, writing, mathematics, or any
other domain of knowledge and skill. Like most other assessments used
in schools, most oral language assessments should be recognized for their
limitations: They do not capture and measure the breath of natural lan-
guage use both in and outside of school, and the evaluative information
they provide about students should not be overinterpreted. Commins and
Miramontes (1989) provide a good demonstration of this in their detailed
ethnographic study of the oral Spanish and English language use of four
5th and 6th Grade native Spanish speakers. Based on general classroom
observations and assessment results, students were judged by their teach-
ers to be lacking conceptual knowledge and full language proficiency in
both languages. In fact, the authors’ qualitative data analyses (200 hours
of classroom observations and 50 hours of observations outside of school)
revealed that students were more proficient in both Spanish and English
than their teachers had been able to determine. Moreover, students showed
higher levels of proficiency in their language use outside of school than
they did in classroom settings. These results were arrived at through far
more extensive and intensive observations and evaluations than classroom
teachers can ever be expected to complete, yet the study offers a caution-
ary tale: even the best oral language assessments and teacher observa-
tions likely sample a small set of language skills from a narrow range of
language tasks and contexts. (See also Gonzalez, Bauerle, and Felix-Holt,
1996.)
P1: PJU
Oral Language 39
Before ending this section, it is important to point out that while it is easy
to criticize oral language assessments on a variety of grounds, developing
better assessments, especially ones that try to capture students’ day-to-day
language abilities, represents a substantial challenge. Work by Gomez et al.
(1996) illustrates this point. Gomez et al. provide a detailed account of their
unsuccessful efforts to develop and validate a language observation scale.
The research team tested their scale against several measurement crite-
ria including interrater reliability, stability over time, internal consistency,
criterion-related validity, and also efficiency. None of the five widely used
observational scales Gomez et al. reviewed prior to their own development
work met any of these criteria, and none had been evaluated for any more
than two of them. Among the several findings they report, Gomez et al.
found that few of the attributes of language use prescribed in the theoreti-
cal literature could be successfully operationalized (seven of twenty-three).
Of the seven attributes that were successfully operationalized, only four
could be observed and evaluated reliably by the research team. In addi-
tion, the research team found it extremely difficult to produce stable scores
over time. Gomez et al. estimate that one would need to collect and ana-
lyze a minimum of six 30-minute language observations over a two- to
three-week period in order to achieve appropriate levels of stability. As
the authors note, such extensive observations are impractical for school
personnel.
conclusions
Studies of language development suggest that question forms develop in
L2 following the same progression observed among monolingual English
children. With increasing proficiency, ELLs demonstrate a wider reper-
toire of question forms, but even less proficient, young ELLs demonstrate
some command over English questions and show considerable growth
over relatively short periods of time (six months to a year). Similarly, with
increasing proficiency, ELLs also demonstrate greater capacity to explicate
what words mean through formal definitions; however, even beginning-
level ELLs, as early as first grade, are able to articulate word meanings
through simple associations or informal definitions. This capacity to expli-
cate word meaning is an example of academic language use. Existing
evidence suggests that the academic uses of language are associated with
higher levels of oral language proficiency and with literacy achievement.
Clearly, however, much more needs to be done to clarify the precise nature
of academic oral language proficiency, independent of literacy and, at the
same time, in relationship to traditional constructs of literacy.
More proficient ELLs also demonstrate a wider repertoire of language-
learning strategies. Those strategies, observable in ELLs as young as
four to five years old, appear to be hierarchical and emerge in the same
P1: PJU
relative order, from receptive strategies to interactive strategies to language

and communication monitoring strategies. At least one study suggests
that explicitly teaching ELLs how to use strategies effectively, especially
metacognitive strategies, might be beneficial, at least for older ELLs. While
it is not possible to generalize based on one study, strategy instruction
has a reasonably strong basis of support in other domains (see Chapter 4).
Thus, it would be important to establish the effects and nature of strategy
instruction as applied to listening and speaking, in particular for ELLs of
different ages and at different levels of proficiency.
Studies of social and personality factors indicate that ELLs can develop
L2 skills without a predisposition to associate with English speakers. In
fact, some level of English proficiency may be necessary before ELLs are
likely to begin associating with English speakers. Moreover, exposure to
English speakers is probably not as critical to language acquisition as the
use of that exposure and the interactions that ensue.
A small set of studies suggests that ELLs require several years of school-
ing to attain L2 oral proficiency. Progress from beginning to middle levels
of proficiency is relatively rapid, but progress from middle to upper levels
of proficiency is slower. Moreover, rates of growth, at least based on exist-
ing studies, appear to be similar for English-speakers learning Spanish
and Spanish-speakers learning English, and more similar than one might
expect for ELLs in programs that vary in terms of their relative emphasis on
English and L1 usage (all-English, two-way, and developmental bilingual).
This finding might be idiosyncratic to this small sample of studies, or it
might be indicative of a reliably constant rate of oral proficiency develop-
ment. Alternatively, it might reflect a general lack of instructional attention
to oral language development once students advance to middle levels of
proficiency.
Clearly, additional research is needed that documents rates of oral pro-
ficiency development. Some may take objection to this proposal insofar as
rates of development or proficiency attainment tend to gloss over consid-
erable individual variation. Nevertheless, with schools under considerable
pressure to establish criteria for adequate yearly progress for ELLs, such
data are critical. States, districts, and schools throughout the country are
trying to define criteria for adequate yearly progress, essentially, without
the aid of any empirically derived estimates.
There is also a clear need for research that examines systematically the
relationship between L2 oral language development and instruction. We
simply do not know empirically the extent to which the rates at which ELLs
achieve oral proficiency can be accelerated. We do not know empirically
the extent to which oral language development is amenable to instruc-
tion, whether defined in terms of direct instruction, natural language
approaches, and/or content-based approaches, all of which are discussed
in the ELL literature and none of which, apparently, has been systematically
P1: PJU
Oral Language 41
investigated in terms of oral outcomes. Given this gap in research, it is not

surprising that teachers tend to express confusion about how best to sup-
port the English oral language development of ELL students (see Gersten
and Baker, 2000).
The majority of studies clustered together under the topic of school fac-
tors focused on ELLs’ classroom language use. These studies suggest that
ELLs are most likely to use the language used to deliver instruction in
their interactions with peers and teachers. Interactive activities that pair
ELLs with native or fluent English speakers can provide language learn-
ing opportunities for ELLs, but careful consideration must be given to the
design of the task, the training of non-ELLs, and the language proficiency of
the ELLs. It also appears that increased L2 use is associated with improved
L2 proficiency. At the same time, several studies imply that while use and
exposure are necessary conditions for language learning, they may not
be sufficient conditions, especially when it comes to achieving higher lev-
els of proficiency involving more academic uses of language. The content
and quality of L2 exposure and use are probably of equal if not greater,
importance than L2 exposure and use per se. Morever, at least one study
(Saville-Troike, 1984) reminds us that classroom language use is a vehicle
for both language and academic development. While the use of L2 may
contribute to L2 proficiency, for students just beginning to acquire English,
the use of L1 may make a stronger contribution to academic development.
According to the research reviewed on non-school factors, there is a
positive relationship between English language use outside of school and
oral English proficiency. However, this finding is qualified by three other
findings. First, although the development of L2 oral proficiency is likely
aided by L2 use outside of school, it is not necessarily impeded by con-
tinued development and use of L1. Second, at advanced levels of English
language and literacy development, English use outside of school may not
be as critical as English use in school. Third, the relationship between L1
and L2 use outside school may be mediated by sociocultural factors: The
emphasis families place on maintaining L1 relative to learning L2 may vary
across cultural and/or SES groups. Unfortunately, these findings on non-
school factors are limited by the small sample of studies on which they
are based. It should be noted, however, that our sample is small primarily
because among the many studies that focus on non-school factors, very
few include measures of oral language development. It seems important
that future research attempt to incorporate such measures in order to estab-
lish empirical links among home, school, and community factors and the
language development of ELLs.
Regarding assessment, most schools in the United States with at least
moderate numbers of ELLs assess oral English language proficiency upon
entry to school and to inform program exit decisions. However, most oral
language instruments are imperfect and produce results that should be
P1: PJU
interpreted cautiously. Most tests in use have not been subjected to rigorous
evaluation; classification cutoffs for proficiency levels can vary from one
test to the next; normative results can be problematic depending on the
match between examinees and the norming group; and non-proficient clas-
sifications can be inaccurate. In short, all oral language assessment results
should be viewed as a sampling of students’ overall oral proficiency. Pre-
dicting success in mainstream classrooms – typically, the process of deter-
mining whether or not to reclassify ELLs as FEP – should involve multiple
sources of information and is likely best served by oral language instru-
ments that are more academically oriented and broad rather than narrow
in scope. A significant qualifier to these findings, however, is the dated-
ness of some of the studies that support them (e.g., Merino & Spencer, 1983;
Ulibarri et al., 1981). However, our search located no recent studies that
evaluated the validity and reliability of more current generations of oral
language assessments. This represents a critical area of needed research.
As a closing remark, this chapter shines a spotlight on an area of the
curriculum – oral language – that typically remains in the shadows. This
has been consistently noted by researchers of L2 development (see Fillmore
and Valadez, 1986), by researchers of L1 development (Loban, 1976), and by
scholars who document the history of the English Language Arts (Squire,
1991). Loban (1976) places blame partly on the fact that oral language is
impervious to paper and pencil assessment and, therefore, has not been
incorporated into the high-stakes testing system that greatly influences
curriculum and research in the United States. The results of our review
confirm what seems to be a continuing neglect of oral-language research.
For the most part, English oral language development for ELLs has over
the last twenty years continued to remain in the shadows of literacy and
mathematics, the mainstays of high-stakes testing.
References
August, D. 1987. Effects of peer tutoring on the second language acquisition
of Mexican-American children in elementary school. TESOL Quarterly 21(3),
717–36.
Baker, K. 1998. Structured English immersion: Breakthrough in teaching limited-
English-proficient students. Phi Delta Kappan 79, 199–203.
Cardoza, D. 1996. The identification and reclassification of limited-English-
proficient students: A study of entry and exit classification procedures. NABE
Journal 11(1), 21–45.
Carlisle, J., Beeman, M., Davis, L., and Spharim, G. 1999. Relationship of metalin-
guistic capabilities and reading achievement for children who are becoming
bilingual. Applied Psycholinguistics 20, 459–78.
Cathcart-Strong, R. L. 1986. Input generation by young second language learners.
TESOL Quarterly 20(3), 515–29.
P1: PJU
Oral Language 43
Chesterfield, R. A., and Chesterfield, K. B. 1985a. Natural order in children’s use of

second language learning strategies. Applied Linguistics 6(1), 45–59.
1985b. “Hoja’s with the H”: Spontaneous peer teaching in bilingual classrooms.
Bilingual Review 12(3), 198–208.
Chesterfield, R. A., Chesterfield, K. B., Hayes-Latimer, K., and Chavez, R. 1983. The
influence of teachers and peers on second language acquisition in bilingual
preschool programs. TESOL Quarterly 17(3), 401–19.
Collier, V. P. 1987. Age and rate of acquisition of second language for academic
purposes. TESOL Quarterly 21(4), 617–41.
Commins, N., and Miramontes, O. 1989. Perceived and actual linguistic
competence: A descriptive study of four low-achieving Hispanic bilingual
students. American Educational Research Journal 26(4), 443–72.
Cummins, J. 1979. Linguistic interdependence and the educational development
of bilingual children. Review of Educational Research 49, 222–51.
Fernandez, M., Pearson, B., Umbel, V., Oller, D., and Molinet-Molina, M. 1992.
Bilingual receptive vocabulary in Hispanic preschool children. Hispanic Journal
of Behavioral Sciences 14(2), 268–76.
Fillmore, L. W., and Valadez, C. 1986. Teaching bilingual learners. In M. Wittrock
(ed.), Handbook of Research on Teaching (pp. 648–85). New York: MacMillan.
Fitzgerald, J. 1995. English-as-a-second-language learners’ cognitive reading
processes: A review of research in the United States. Review of Educational
Research 65(2), 145–90.
Garcia-Vázquez, E., Vasquez, L. A., Lopez, I. C., and Ward, W. 1997. Language
proficiency and academic success: Relationships between proficiency in two
languages and achievement among Mexican-American students. Bilingual
Research Journal 21(4), 395–408.
Genesee, F. (ed.). 1999. Program alternatives for linguistically diverse students
(Educational Practice Report 1). Santa Cruz, CA, and Washington, DC: Center
for Research on Education, Diversity, and Excellence.
Gersten, R., and Baker, S. 2000. The professional knowledge base on instructional
practices that support cognitive growth for English-language learners. In
R. Gersten, E. Schiller, and S. Vaughn (eds.), Contemporary Special Education
Research: Syntheses of the Knowledge Base on Critical Instructional Issues (pp. 31–
79). Mahwah, NJ: Lawrence Erlbaum Associates.
Goldstein, B., Harris, K., and Klein, M. D. 1993. Assessment of oral storytelling
abilities of Latino junior high school students with learning handicaps. Journal
of Learning Disabilities 26(2), 138–43.
Gomez, L., Parker, R., Lara-Alecio, R., Ochoa, S. H., and Gomez, R. 1996. Naturalistic
language assessment of LEP students in classroom interactions. The Bilingual
Gonzalez, V., Bauerle, P., and Felix-Holt, M. 1996. Theoretical and practical impli-
cations of assessing cognitive and language development in bilingual children
with qualitative methods. The Bilingual Research Journal 20(1), 93–131.
Hakuta, K., Butler, Y. G., and Witt, D. 2000. How long does it take English learn-
ers to attain proficiency? Linguistic Minority Research Institute (ERIC Doc-
ument Reproduction Service No. FL 026 180). (Available at LMRI website:
http://www.lmri.ucsb.edu/resdiss/pdf˙files/hakuta.pdf).
P1: PJU
Hansen, D. A. 1989. Locating learning: Second language gains and language use
in family, peer and classroom contexts. NABE Journal, 13, 161–79.
Howard, E. R., Christian, D., and Genesee, F. 2003. The development of bilingualism
and biliteracy from grade 3 to 5: A summary of findings from the Cal/CREDE study
of two-way immersion education. Santa Cruz, CA, and Washington, DC: Center
for Research on Education, Diversity, and Excellence.
Jacob, E., Rottenberg, L., Patrick, S., and Wheeler, E. 1996. Cooperative learning:
Context and opportunities for acquiring academic English. TESOL Quarterly
30(2), 253–80.
Johnson, D. M. 1983. Natural language learning by design: A classroom experiment
in social interaction and second language acquisition. TESOL Quarterly 17(1),
55–68.
Krashen, S. 1996. A gradual exit, variable threshold model for limited-English-
proficient children. NABE News 19(7), pages 1 and 15–17.
Lambert, W. E., and Taylor, D. M. 1996. Language in the lives of ethnic minorities:
Cuban-American families in Miami. Applied Linguistics 17, 478–500.
Lindholm, K. J. 1987. English question use in Spanish-speaking ESL children:
Changes with English language proficiency. Research in the Teaching of English
21(1), 64–91.
Lindholm-Leary, K. J. 2001. Dual Language Education. Clevedon, UK: Multilingual
Matters.
Loban, W. 1976. Language development: Kindergarten through grade twelve (Research
Report No. 18). Urbana, IL: National Council of Teachers of English.
MacSwan, J., Rolstad, K., and Glass, G. 2002. Do some school-age children have
no language? Some problems of construct validity in the Pre-LAS Español.
Bilingual Research Journal 26(2), 395–419.
Malave, L. 1989. Contextual elements in a bilingual cooperative setting: The expe-
riences of early childhood LEP learners. NABE Journal 13, 96–122.
Medina, M., and Escamilla, K. 1992. English acquisition by fluent- and limited-
Spanish-proficient Mexican-Americans in a 3-year maintenance bilingual pro-
gram. Hispanic Journal of Behavioral Sciences 14(2), 252–67.
Merino, B., and Spencer, M. 1983. The comparability of English and Span-
ish versions of oral language proficiency instruments. NABE Journal 7(2),
1–31.
Milk, R. D. 1982. Language use in bilingual classrooms: Two case studies. In M.
Hines and W. Rutherford (eds.), On TESOL ‘81 (pp. 181–91). Washington, DC:
TESOL.
O’Malley, J. M., Chamot, A. U., Stewner-Manzanares, G., Kupper, L., and Russo,
R. P. 1985a. Learning strategies used by beginning and intermediate ESL stu-
dents. Language Learning 35(1), 21–46.
O’Malley, J. M., Chamot, A. U., Stewner-Manzanares, G., Russo, R., and Kupper,
L. 1985b. Learning strategy applications with students of English as a second
language. TESOL Quarterly 19(3), 557–84.
Pease-Alvarez, L. 1993. Moving in and out of bilingualism: Investigating native language
maintenance and shift in Mexican-descent children (Research Report No. 6). Santa
Cruz, CA: University of California, National Center for Research on Cultural
Diversity and Second Language Learning.
P1: PJU
Oral Language 45
Pease-Alvarez, L., and Winsler, A. 1994. Cuando el maestro no habla Espanol:

Children’s bilingual language practices in the classroom. TESOL Quarterly
28(3), 507–35.
Peck, S. 1987. Signs of learning: Child nonnative speakers in tutoring sessions with
a child native speaker. Language Learning 37(4), 545–71.
Platt, E., and Troudi, S. 1997. Mary and her teachers: A Grebo-speaking child’s place
in the mainstream classroom. The Modern Language Journal 81, 28–49.
Rodriguez-Brown, F. V. 1987. Questioning patterns and language proficiency in
bilingual students. NABE Journal 13(3), 217–33.
Royer, J. M., and Carlo, M. S. 1991. Assessing the language acquisition progress of
limited English proficient students: Problems and a new alternative. Applied
Measurement in Education 4(2), 85–113.
Saville-Troike, M. 1984. What really matters in second language learning for aca-
demic achievement? TESOL Quarterly 18(2), 199–219.
1988. Private speech: Evidence for second language learning strategies during
the “silent” period. Journal of Child Language 15(3), 567–90.
Schecter, S., and Bayley, R. 1997. Language socialization practices and cultural
identity: Case studies of Mexican-descent families in California and Texas.
TESOL Quarterly 31(3), 513–41.
Schrank, F., Fletcher, T. V., and Alvarado, C. G. 1996. Comparative validity of three
English oral language proficiency tests. The Bilingual Research Journal 20(1),
55–68.
Snow, C. E., Cancino, H., Gonzalez, P., and Sriberg, E. 1987. Second language learners’
formal definitions: An oral language correlate of school literacy (Tech. Rep. No. 5).
Los Angeles: University of California, Center for Language Education and
Research.
Squire, J. 1991. The history of the profession. In J. Flood, J. Jensen, D. Lapp, and
J. Squire (eds.), Handbook of research on teaching the English language arts (pp. 3–
17). New York: Macmillan.
Strong, M. 1983. Social styles and the second language acquisition of Spanish-
speaking kindergartners. TESOL Quarterly 17(2), 241–58.
1984. Integrative motivation: Cause or result of successful second language
acquisition? Language Learning 34(3), 1–13.
Thomas, W., and Collier, V. 2002. A national study of school effectiveness for language
minority students’ long-term academic achievement final report: Project 1.1 Santa
Cruz, CA, and Washington, DC: Center for Research on Education, Diversity,
and Excellence.
Ulibarri, D. M., Spencer, M. L., and Rivas, G. A. 1981. Language proficiency and
academic achievement: A study of language proficiency tests and their rela-
tionship to school ratings as predictors of academic achievement. NABE Journal
5(3), 47–79.
Umbel, V. M., and Oller, D. K. 1994. Developmental changes in receptive vocabulary
in Hispanic bilingual school children. Language Learning 44(2), 221–42.
Umbel, V., Pearson, B., Fernandez, M., and Oller, D. 1992. Measuring bilingual
children’s receptive vocabularies. Child Development 63(4), 1012–20.
Weslander, D., and Stephany, G. V. 1983. Evaluation of English as a second language
program for Southeast Asian students. TESOL Quarterly 17(3), 473–80.
P1: PJU
46
0521859751c02a
table a.2.1. Summary of Studies on Language Development
Sample Instructional Outcome

Authors Characteristics Grades Method or Program Design Measures Results
Carlisle et al. Mexican- 1, 2, 3 English-only Within group: – PPVT: Spanish and – Significant
(1999) American ELLs, parochial school Correlational English receptive correlation only for
low SES, low vocabulary Spanish-English
CB989B/Genesee
– n = 57
achieving – Formal and informal formal definitions
definitions
Chesterfield Mexican- Pre-K-1 Bilingual program Within group: – Observed use of 12 – Strategies
and American ELLs Longitudinal language learning hierarchically
Chesterfield from Spanish- – n = 14 strategies organized
0 521 85975 1
(1985a) speaking homes – Higher levels of L2

in a border Texas oral proficiency
city associated with
more varied
strategies
Garcia- Bilingual Hispanic 6–12 Not reported Within group: – WLPB: English oral – Significant
Vázquez students Correlational proficiency correlation
et al. (1997) – n = 100 – ITBS: English reading between English
achievement oral proficiency
and reading
Goldstein, Mexican- 7–9 L1 instruction Within group: – English story retell – Moderate, sign
Harris, and American, fluent through Correlational scored for lang and correlation for
Klein (1993) English proficient, Grade 3 – n = 31 content scores retell content and
previously ELL (adapted from reading
November 12, 2005
LAS-O) – No sign correlation

– PIAT: English reading for lang and
15:5
reading
P1: PJU
0521859751c02a
Lindholm Mexican- K, 2 Varying involvement Between-group: – Question forms in – All children

(1987) American from in bilingual longitudinal English observed showed growth in
Spanish-English program – Limited during home visits question use over
bilingual families English prof over one-year period one-year period
(n = 2) – Fluent English prof
– Fluent English students showed
CB989B/Genesee
prof (n = 2) command of wider

– One K and one array of question
2nd in each forms
pair
O’Malley, Spanish- and High Listening and Random Pre- and post-testing: – TR2 > CO on
Chamot, Vietnamese- school speaking strategy assignment to: – Listening to lecture post-test speaking
0 521 85975 1
Stewner- speaking ELLs, training – TR1: metacog followed by (oral presentation)

Manzanares, in intermediate + cognitive comprehension test – No sign difference
Kupper and level ESL classes strategies – Oral presentation, on listening
Russo (n = 27) videotaped and
(1985a) – TR2: cognitive scored by research
strategies team
(n = 26)
– CO: no
strategy
training
(n = 22)
(continued)
November 12, 2005
15:5
47
P1: PJU
48
0521859751c02a
table a.2.1 (continued)

O’Malley, Spanish- and High Beginning or Within group: – Self-reported use of – Metacognitive,
Chamot, Vietnamese- school intermediate ESL Numerical language learning cognitive, and
CB989B/Genesee
Stewner- speaking ELLs, results, no stat strategies social strategies

Manzanares, in beginning or analysis identified
Russo, and intermediate – n = 70 – Students tended to
Kupper level ESL classes use more cog than
(1985b) metacog strategies
0 521 85975 1
– More likely to use

cog strategies on
less rather than
more demanding
tasks
Rodriguez- Hispanic, fluent 3 Two-way immersion Between group: – Question forms in – Students showed
Brown Spanish-limited English and English and Spanish command of wider
(1987) English speakers Spanish observed in and out array of question
and fluent speakers; of school forms in their
English-limited some stronger stronger language
Spanish speakers in Spanish,
some stronger
in English
– numeric
November 12, 2005
results, no stat
analysis
–n=6
15:5
P1: PJU
0521859751c02a
Royer and Puerto 5–6 Transitional bilingual Within group: – Sentence verification – No significant
Carlo (1991) Rican-American program Longitudinal technique: English correlation
ELLs, – n = 49 listening and reading between English
transitioning from comp listening and
Spanish to reading at either
English grade 5 or 6
CB989B/Genesee
instruction
Saville-Troike L1 and ethnicity 2–5 English-only with Within group: – English reading – Moderate,
(1984) varied, first year ESL; 30 min/day of Correlational – Several measures of significant
in United States, L1 – n = 19 oral English correlation with
parents in proficiency reading only for
graduate program – Classroom language measure of
0 521 85975 1
for foreign use vocabulary

students
Saville-Troike Chinese, Japanese, Pre-K-3 English-only Ethnographic – Intrapersonal learning – Six of nine
(1988) Korean speakers, programs description strategies exhibited in experienced “silent
first year in and transcript private speech period;” all six
United States, excerpts evidenced
parents in –n=9 strategies
graduate program
for foreign
students
(continued)
November 12, 2005
15:5
49
P1: PJU
50
0521859751c02a

CB989B/Genesee

Snow et al. Mid SES ELLs and 2, 3, 4, 5 ESL for ELLs; Between group – Quality of informal – Quality of formal
(1987) English speakers French for English comparisons: and formal def. varied by prof
learning French speakers – Quality of definitions of – Sign correlation
0 521 85975 1
definitions by common words between formal def

low and high and reading
English prof – Magnitude of
– Correlations correlation
by grade level stronger in
– n = 137 grades 4 and 5 than
in grade 2
Strong (1983) Span-speaking K Bilingual class with Within group: – Natural – Some social styles
ELLs enrolled in ELLs, and native correlational communicative lang: correlated w/ some
one classroom, and fluent English – Between English voc, syntax, lang measures
small industrial speakers group pronunciation; social – Successful ELLs
community in comparisons: styles (e.g., made good use of
California more and less talkativeness); contact with
November 12, 2005
successful L2 contacts with English English speakers

learners speakers
– n = 13
15:5
P1: PJU
0521859751c02a
Strong (1984) Span-speaking K Bilingual class with Within group: – Nominations of – High English prof
ELLs ELLs, and native Correlational English speaker more likely to
and fluent English – Between gp playmates nominate English
speakers comparisons: – English voc, syntax, speakers as
low and high pronunciation playmates
English prof – Low English
CB989B/Genesee
– n = 19 prof increased
English speaker
as their prof
increased
Ulibarri, 1,110 Hispanic 1, 3, 5 Not reported Between group: – LASO, BSM, BINL: – Stronger
Spencer, students Correlational Oral L2 language use correlations with
0 521 85975 1
and Rivas representing four – grade 1 – Standard scores from reading at all
(1981) major regions of (n = 165) either CAT, CTBS, or grades for broader,
California: Bay – grade 3 SAT (district’s chosen more academic
Area, Central (n = 196) assessment): reading measures of oral
Valley, Los – grade 5 prof-LAS-O
Angeles, San (n = 207)
Diego
November 12, 2005
15:5
51
P1: PJU
52
0521859751c02a
table a.2.2. Summary of Studies on School Factors

CB989B/Genesee
August (1987) Mexican- 1–5 6-week Btwn grp: – Observations of – TR sign > CO in
American ELLs peer-tutoring Matched quantity and type of frequency of use of
from Spanish- training pairs, random verbal interactions English
speaking homes, assign – PPVT: vocab – Sign correlation
low SES, – group diff – LAS-O: oral prof between amount of
multiethnic – correlations – Pre- and post- English and PPVT
0 521 85975 1
community – n = 12 measures and also with some

LAS-O subtests
Cathcart-Strong Spanish-speaking K ELLs in a bilingual Within group: – Response pattern – English-speakers
(1986) ELLs and program descriptive among English- did not provide
English-speakers – n = 4 ELLs speakers to ELLs in worthwhile
at one school in play situations interaction to ELLs
suburb of San
Francisco
Chesterfield Mexican- 1 All-English and Btwn group: chi – Language use during – ELLs in all-English
and American, Spanish-English squares peer interactions and used sign more
Chesterfield Spanish- bilingual – All-English oral proficiency English than
(1985b) speaking ELLs programs classes Spanish;
from 11 (n = 10) – ELLs in bilingual
November 12, 2005
classrooms in – Bilingual used sign more

California and classes Spanish than
Texas (n = 10) English
15:5
P1: PJU
0521859751c02a
Chesterfield Spanish-speaking Pre-K Bilingual program Within group: – ELLs use of English – Significant positive
et al. (1983) ELLs from low- Correlational and Spanish rank-order
income families – n = 11 – Oral English correlations
in Texas and proficiency (mean between increase in
Wisconsin length of utterance) English use and
measured via increases in L2 oral
CB989B/Genesee
observations at proficiency
beginning and end of
year
Hakuta, Butler, ELLs of various 1, 2, 3, 4, English-only Cross-sectional: – IPT: Oral English – Almost all English
and Witt L1s enrolled in 5 program from one oral prof proficiency proficient by
(2000) same district district across grades grade 5
0 521 85975 1
Sample A since K, – All ELLs with – Largest gain from

classified as ELL available data grades 1 to 2
in K, Northern included
California – n = 1,872
Hakuta, Butler, Spanish-speaking 1, 3, 5 Some from Cross-sectional: – Oral English – Gains evident from
and Witt ELLs from high- bilingual, some Oral prof proficiency (WLPB-R) grades 1 to 5
(2000) poverty schools English-only across grades – 3 + years behind
Sample B enrolled in same – About 40 ELLs native English
district since K, randomly speaker norms at
classified as ELL selected per grade 5
in K, Northern grade
California – n = 122
November 12, 2005
(continued)
15:5
53
P1: PJU
54
0521859751c02a
CB989B/Genesee

Howard, Low SES Hispanic 3–5 Two-way Longitudinal: – Parallel, – Almost all English
0 521 85975 1
Christian, and Spanish- immersion Oral prof study-developed proficient by

Genesee speaking ELLs & program across grades measures of oral grade 5
(2003) mid SES – Spanish- Spanish & English – Similar rates of L2
English-speakers speakers (n = proficiency oral prof growth
approx. 125) for native Spanish
– English- and native English
speakers speakers
(n = 115)
Jacob et al. Limited and fluent 6 Coop. learning; Within group: – Observed instances of – Few language
(1996) proficient ELLs mainstream Social – Numeric potential language learning
in one ethnically Studies class results, no stat learning opportunities opportunities
mixed analysis – Most (54%)
classroom, – Transcript involved clarifying
November 12, 2005
working class excerpts pronunciation

Eastern United –n=7 during oral
States reading
15:5
P1: PJU
0521859751c02a
Johnson (1983) Mexican- K-4 5-week training: Btwn gp: – Observations of – TR sign > CO at
American ELLs Inter-ethno- matched quantity and type of post on quantity of
from Spanish- linguistic Peer pairs, random verbal interactions interaction with
speaking homes Tutoring (IEPT) assign: – PPVT: Vocab fluent English
participating in a – TR: IEPT – LAS – O: Oral prof tutors
summer school – CO: No – Child-Child – TR sign > CO at
CB989B/Genesee
program at training, Communication Test post on PPVT, but

Stanford – ELLs (n = 16) (CCCT) not LAS-O or
experimental – Fluent – Pre- and post- CCCT
school English- measures
speaking
tutors (n = 18)
0 521 85975 1
Lindholm- Hispanic, Spanish- K, 1, 2, Dual language Cross-sectional: – Student Oral Language – Almost all English
Leary speaking ELLs, 3, 4, 5 (two-way) Oral prof Observation Matrix proficient by
(2001) Hispanic-, immersion across grades (SOLOM): Teacher grade 5
European-, program, 90/10 – Spanish- ratings of students’ – Similar rates of L2
African-, and model speakers oral language oral prof growth
Asian-American (n = 2,938) proficiency for native Spanish
English- – English- and native English
speakers, high speakers speakers
and low ethnic (n = 1,715)
density and SES
(continued)
November 12, 2005
15:5
55
P1: PJU
56
0521859751c02a

Malave (1989) Hispanic, Spanish- 2 Bilingual program Within group: – Observed verbal – Although
speaking ELLs Numeric behaviors and mid-level English
with mid-level results, no stat language choice during prof, on average,
CB989B/Genesee
English prof, analysis small group activities ELLs used Spanish

low-mid SES, –n=5 87% of time
western New
York
Milk (1982) Spanish-speaking 2 “Model” Between – Observed speech acts – Spanish dominant
ELLs from maintenance groups: and language choice and balanced
0 521 85975 1
Spanish- bilingual program Numeric during small group bilinguals rarely

speaking homes, results, no stat activities used English
both Spanish- analysis
dominant and – Spanish dom
balanced (n = 8)
bilinguals – Balanced bil
(n = 8)
– English dom
(n = 8)
Medina and Spanish-speaking K-2 Developmental – Longitudinal: – LAS-O in English and – Stronger oral gains
Escamilla ELLs from 24 (maintenance) Oral prof at K Spanish: English and for all ELLs in
(1992) schools in bilingual program and grade 2 Spanish oral English than
Arizona school – ELLs (n = 298, proficiency among limited
November 12, 2005
district of whom 187 Spanish prof in

limited Spanish
Spanish prof)
15:5
P1: PJU
0521859751c02a
Pease-Alvarez U.S.-born 4 Bilingual program Case study: – Spanish and English – Significant shift
and Winsler Mexican- and transitioning Multi subject, use at school with over year from
(1994) American, from Spanish to numeric teacher, peers, self Spanish to English
Spanish- English results, no stat use for all three
speaking ELLs, instruction analysis ELLs
immigrant, –n=3
CB989B/Genesee
working-class
parents
Peck (1987) Mexican and K ELLs tutored by Between – ELL language output – Language outputs
Mexican- trained Grade 2 groups: observed during consistent with
American English speaker Chi-squares tutoring sessions proficiency levels:
Spanish- – hi (n = 3) H&M>L
0 521 85975 1
speaking ELLs – med (n = 3)

from three – low (n = 3)
classrooms at – Selected based
one school in LA on obs to max
area L2 variation
Platt & Troudi Liberian, 3 English mainstream Single subject – Daily, year-long – Assistance from
(1997) Grebo-speaking classroom case study: observations of child’s English-speakers
girl, first year in descriptive interactions, and not sufficient to
United States –n=1 language proficiency meet ELL’s needs
– minimal lang
growth
(continued)
November 12, 2005
15:5
57
P1: PJU
58
0521859751c02a

Saville-Troike Ethnicity, L1 2, 3, 4, 5 English-only with Within group: – English reading – Significant
(1984) varied, 1st year ESL; 30 min/day correlational – Several measures of correlation
in United States, L1 – n = 19 oral Eng proficiency between L2 use
CB989B/Genesee
parents in grad – Classroom language and L2 prof

program for use – No correlation
foreign students between use and
rdg
Thomas & Mostly Hispanic, 1, 2, 3, 4, Two-way dual – Quasi -SOLOM: Teacher’s – Almost all English
0 521 85975 1
Collier (2002) low SES, Spanish 5 language longitudinal ratings of students Eng proficient by
L1, and English immersion (Data avail for and Span oral grade 5
L1, Northwest multi grades proficiency – Similar rates of L2
United States for most oral prof growth
students) for native Spanish
– Spanish- and native English
speakers speakers
(n = 238)
– English-
speakers
(n = 279)
Weslander & Southeast Asians, 2–10 District developed Within group: – BSM: Oral proficiency – Average 1-year
Stephany in United States ESL program for pre/post test measure gain about half a
(1983) for 2–68 months, recent immigrants – n = 577 –Taken in fall and spring proficiency level
November 12, 2005
Des Moines,
Iowa
15:5
P1: PJU
0521859751c02a
table a.2.3. Summary of Studies on Non-School Factors

Authors Characteristics Grades Program Design Measures Results
Hansen (1989) Mexican-Am 2, 5 Bilingual Within group: – Stanford – Positive
ELLs, Spanish programs correlational Diagnostic Reading relationship
CB989B/Genesee
dominant (regression) Test: Auditory between English

homes, San – n = 117 vocabulary use at home and
Francisco area English vocabulary
Lambert and Cuban-American Not Not reported Between group: – Mothers’ – Working-class
Taylor (1996) mothers living in applicable middle- and self-reports: mothers tend to
0 521 85975 1
Miami with working-class language use, emphasize English

middle-school- mothers attitudes toward – middle-class
age children – n = 108 (56 per gp) Spanish and mothers tend to
English, ratings of emphasize
child’s proficiency maintenance of
Spanish
Pease-Alvarez Mexican- 3 Bilingual Between group: 4 – PPVT, TVIP-H: – Positive
(1993) American ELLs programs gps ranging from Spanish and relationship
and their born in Mexico/ English oral between English
families living in use mostly Spanish vocabulary use at home and
California at home to born in – Parents self-reports English vocab
United States/use of language use in – Negative
mostly English at home relationship
home between English
November 12, 2005
– n = 55 (n varies per use at home and

gp) Spanish vocab
15:5
59
(continued)
P1: PJU
60
0521859751c02a

CB989B/Genesee
Schecter and Mexican- Varies, but Not reported Case study of each – Parents’ – Maintaining
Bayley (1997) American all children family descriptions of Spanish tied to
families living in elementary – n = 4 families efforts to support cultural identity
California and school age English and – Maintenance
Texas maintain Spanish strategies varied
across families
0 521 85975 1
Umbel and Cuban-American, 1, 3, 6 English Between group: 2 – PPVT, TVIP-H: – Equal Spanish-
Oller (1994) bilingual immersion gps – more Spanish Spanish and English at home
parents, children programs than English and English oral grp scored higher
introduced to Spanish-English vocabulary on English vocab
English and used equally at than more
Spanish since home Spanish than
birth – n = 102 (44 per gp) English gp
(simultaneous – No gp diffs on
bilinguals) Spanish vocab
– Spanish vocab
sign 1st order
predictor of
English vocab
November 12, 2005
15:5
P1: PJU
0521859751c02a
table a.2.4. Summary of Studies on Assessment

Cardoza School districts Pre-K-12 75% of districts Survey: Districts – Survey responses: – Most schools assess
(1996) from 10 states offered some selected from target what does the ELLs’ oral Eng prof
CB989B/Genesee
with highest form of states randomly district assess for at entry and exit
number of ELLs bilingual – n = 93 ELLs, when do – Most common:
programming they do it, and LAS, BSM, LAB
with what
instruments?
Commins & Underachieving 5, 6 Bilingual Case study: – Researcher – Students
0 521 85975 1
Mira- ELLs in program descriptive observations of demonstrated

montes innovative designed for –n=4 students’ Span and greater proficiency
(1989) program, underachievers Eng use across than school had
presumed to be school and presumed, and
limited Span and non-school settings higher levels of
Eng proficiency proficiency in
non-school settings
than in classroom
settings
Fernandez – Study sample: Preschool Not reported Between group: – PPVT and TVIP-H: – Order of difficulty
et al. Mid-level SES Study sample and Eng and Span of words on Span
(1992) bilingual Hisp, in norm group receptive measure for study
Miami – n = 396 (study vocabulary sample differed
November 12, 2005
– Norm group: sample) from that of the

mid-level SES norm group
monolingual
15:5
Hisp, in Puerto
61
Rico and Mexico (continued)
P1: PJU
62
0521859751c02a

Gomez Bilingual Hisp Pre-K – 6 Students had Validity and – Social language – Scale failed to meet
et al. students in been enrolled reliability study: proficiency (e.g., most measurement
(1996) summer in either Naturalist lang topic development, criteria and did not
program bilingual or observational scale understandability, meet efficiency
ESL program – n = 24 absence of criteria
CB989B/Genesee
hesitations)
Gonzalez, 1st, 2nd, & 3rd K Either bilingual Validity and – Qualitative Use of – Gifted best
Bauerle, generation or English reliability study: English and identified via
and Mexican- program Qualitative scale Spanish Tasks multiple verbal
Felix-Holt American, for identifying (QUEST), includes and nonverbal
0 521 85975 1
(1996) mostly gifted bilingual verbal (Span & Eng) measures in both
Span-speaking children and nonverbal L1 and L2
– n = 17 classification tasks
MacSwan, Span-speaking Pre-K, K Not reported Critical validity – Pre-LAS subtests: – Classifications
Rolstad, ELLs in a large study of Pre-LAS acting out weighted heavily
and Glass metropolitan (Language commands, naming on last two
(2002) school district Assessment Scales) items in pictures, components
classified as – n = 6,118 matching pictures – At least 51% likely
non-Spanish- to phrases, sentence misclassified as
proficient (non- repetition, sentence non-Spanish-
Spanish- completion, story proficient (all
speaking: retell passed
NSS) components 1–4)
Merino and Not applicable Not Not applicable Critical review of – LAS, BSM, LAB, – None of the
November 12, 2005
Spencer applicable technical BINL, and BOLT: assessments

(1983) information oral language provided adequate
15:5
reported by test assessment technical

publishers information
P1: PJU
0521859751c02a
Schrank, Span-Eng bilingual K, 2 Not reported Validity study of oral – 3 assessments: IPT-I, – Strong inter-test
Fletcher, children from language LAS (Pre-LAS & correlations among
and Arizona and assessments LAS-O), & WLPB-R all assessments
Alvarado Texas – n = 196 – Teacher ratings of – WLPB-R correlated
(1996) students’ academic most strongly with
language teachers’ ratings of
CB989B/Genesee
proficiency academic language

(Houston school proficiency
district rating scale)
Ulibarri, Hispanic students 1, 3, 5 Not reported Validity study of oral – LAS-O, BSM, BINL: – Assessments did
Spencer, representing four language Oral language not classify
and Rivas major regions of assessments assessment students
0 521 85975 1
(1981) California: Bay – n = 1,000 – Standard scores for consistently

Area, Central reading & math ANOVA: sign diffs in
Valley, Los from CTBS, CAT, achievement based
Angeles, San SAT (depending on on LAS-O
Diego district): classifications, not
achievement for BSM or BINL
– Teacher ratings of Stepwise
achievement, Regression:
language – Teacher ratings the
proficiency, best 1st-order
likelihood of predictor of
success in achievement, but
mainstream LAS-O was a sign
November 12, 2005
program 2nd- or 3rd-order

predictor
– BSM and BINL
15:5
were not
63
P1: JZZ
Literacy
Crosslinguistic and Crossmodal Issues
Caroline Riches and Fred Genesee
introduction
Crosslinguistic relationships between the L1 and the L2 as well as cross-
modal relationships between oral and written language provide a basis
for discussing research on the reading and writing development of ELLs.
There are two fundamental and inescapable reasons why this is so. First of
all, the learners under consideration, by definition, are acquiring literacy in
English as a second language and have an ongoing developmental history
in their first language. As a result, the relationship between their L1 and
their L2 figures prominently in much of the research on reading and writing
development in ELLs. In fact, relatively little research looks at L2 literacy
development in ELLs without reference to their L1. Second, since reading
and writing in any language implicate both oral and written modes of lan-
guage, the relationship between oral and written language in the L1 and L2
of ELLs has also been a primary theme in much of the research reviewed
here. The questions are: What is the relationship between oral and writ-
ten language development? Is it the same for native English speakers and
ELLs? The following specific relationships are examined in the sections
that follow:
1. L1 oral proficiency and the development of L2 literacy,

2. L2 oral proficiency and the development of L2 literacy,
3. Specific component skills or abilities related to oral and written lan-
guage and the development of L2 literacy, and
4. L1 literacy and the development of L2 literacy.
The crosslinguistic and crossmodal relationships identified herein are com-

plex and interwoven. Consequently, a number of the studies reviewed
focused on more than one of these issues. For example, Lanauze and Snow
(1989) and Langer, Barolome, and Vasquez (1990) examined the relation-
ship between L1 and L2 literacy development and considered aspects of
64
P1: JZZ
Literacy 65
both L1 and L2 oral proficiency. Studies such as these are discussed in all
of the relevant subsections.
theoretical background
A number of theoretical perspectives have served as the impetus or start-
ing point for many of the studies reviewed here and these warrant some
consideration before proceeding with our synopsis. The developmental
interdependence hypothesis (Cummins 1981, 1991) recurs frequently in
many studies (see MacSwan and Rolstad, 2003, for a critique of Cum-
mins’s hypothesis). This hypothesis defines the nature of the relationship
between the L1 and the L2 of ELLs and in so doing distinguishes different
types of language proficiency. On the one hand, some language skills are
fundamentally interpersonal in nature and characterize everyday social
conversations and usage. These skills are thought to be language spe-
cific. Interpersonal communication is usually embedded in meaningful
and immediate contexts so that the meanings participants seek to convey
are supported by shared context and, moreover, the participants are able
to actively negotiate meaning directly through feedback to one another.
Context may be shared by virtue of common past experiences or by the
immediate setting in which communication is taking place. A face-to-face
conversation about a movie that two people have seen is an example of
context-embedded communication. The speakers have the shared experi-
ence of seeing the movie that they can draw on when discussing it. Talking
about a football game that both speakers are watching is another example of
context-embedded communication. These language skills are often impli-
cated in oral uses of language, although not necessarily, and are acquired
relatively quickly in the first language of all normal children.
On the other hand, other language skills serve more complex cognitive
or academic purposes and are characteristically used in decontextualized
ways, such as during educational instruction. Decontextualized language
skills are often associated with written forms of language (e.g., reading a
history textbook), but not necessarily since they can also occur during oral
language use – such as during a lecture on history. Because the conversation
takes place without an immediate or explicit context, care must be taken
to provide context and details that will make the message meaningful.
Returning to our movie example, if the conversational partners are talking
about a movie that they have not both seen, then background information
or “framing” is required on the part of the person who has seen it to make
sure that his or her meaning is clear and to explicate personal views about
the movie because the participants have little recourse to immediate con-
textual cues or shared experiences to draw on. As Cummins notes (2000, pp.
68–9): “the essential aspect of academic language proficiency [underlining
added by authors] is the ability to make complex meanings explicit . . . by
P1: JZZ
66 Caroline Riches and Fred Genesee
means of language itself rather than by means of contextual support or par-

alinguistic cues (e.g., gestures, intonation etc.).” Indeed, success in school
in the long run ultimately depends critically on students’ ability to read
about or express abstract, complex ideas without the benefit of past expe-
rience or concurrent contextual cues.
Academic language proficiency is posited to be part of “a common
underlying proficiency” made up of knowledge and abilities that once
acquired in one language are potentially available for the development of
another (Lanauze and Snow, 1989; Royer and Carlo, 1991). Literacy-related
abilities fall into this latter category. When one has learned to read, then
there are many components of reading that can be used in learning to
read another language. While interpersonal communication skills and the
language abilities that underlie the use of language in contextualized situ-
ations are often acquired relatively rapidly in a second language, research
suggests that more time is needed to acquire proficiency in an L2 for aca-
demic and decontextualized uses – it is reported that five years or more
may be required for ELLs to develop proficiency in English as a second
language for academic purposes that is comparable to that of same-age
native speakers (Collier, 1987; Cummins, 1981, 1992; Lindholm and Aclan,
1991).
An additional related theoretical construct that has been addressed in
this corpus is the threshold hypothesis (Cummins, 1979, 1981; Toukomaa
and Skutnabb-Kangas, 1977; for example, Lindholm and Aclan, 1991,
examined in this issue). The threshold hypothesis posits that both lan-
guage and cognitive development are enhanced if certain levels and types
of proficiency are attained in either or both the L1 and the L2. In con-
trast, linguistic and cognitive enhancement are less like to occur if learners
acquire limited competence in one or both languages. Together, the inter-
dependence and threshold hypotheses raise a number of theoretically and
pedagogically important developmental issues concerning the crosslin-
guistic and crossmodal aspects of language and their crisscrossing effects
on bi- and multilingual development. These issues continue to be at the
forefront of research into the development of literacy in bilingual settings
(Cummins, 1997).
Echoing a contrastive analysis framework (Lado, 1957), some studies in
this corpus have examined differences and similarities between ELLs’ L1
and L2 and their effects on the development of reading and writing abilities
by ELLs. According to the Contrastive Analysis Hypothesis, learners will
have difficulty acquiring aspects of a second language that differ from their
first language, but their acquisition of a second language will be facilitated
by similarities between their first and second languages. A contrastive
analysis perspective is evident, for example, in studies that have examined
similarities and differences in sound – letter correspondences in the L1
and the L2 and their effects on L2 writing development (e.g., Fashola et al.,
P1: JZZ
Literacy 67
1996; Zutell and Allen, 1988) and the effect of crosslinguistic cognates on
vocabulary development (e.g., Hancin-Bhatt and Nagy, 1994; Nagy, Garcia,
Durgunoglu and Hancin-Bhatt, 1993). These effects are commonly referred
to as positive and negative transfer. A number of studies in this review have
sought to identify instances of positive and negative transfer from the L1
during L2 literacy development.
Other articles in this corpus are based on interlanguage principles (e.g.,
Cronnell, 1985; Tompkins, Abramson, and Pritchard, 1999). Interlanguage
theory postulates that second-language acquisition is dynamic and char-
acterized by a series of intermediary stages, from early to advanced, that
reflect influences from the L1 and from developmental processes associ-
ated with the target L2. For example, Tompkins, Abramson, and Pritchard
(1999) identified patterns of L2 development that were similar to those
of English L1 learners. Such effects are commonly referred to as devel-
opmental because they reflect developmental patterns that characterize
native speakers of the language in question. Interlanguage theory is one
of a group of contemporary theories that emphasize cognitive strategies in
L2 acquisition (see Chapter 3 in Ellis, 1986, for an overview).
We now turn to a review of research related to each of these develop-
mental conceptualizations. We have included tables highlighting pertinent
details of the studies at the end of the chapter.
l1 oral proficiency and l2 literacy

Much contemporary theory on literacy education emphasizes the need to
draw on students’ sociocultural experiences (e.g., Heath, 1983; Hudelson,
1994; Maguire and Graves, 2001) and their preexisting knowledge about
reading and writing, including emergent literacy (e.g., Sulzby and Teale,
1991) as a basis for the development of initial literacy abilities in school. The
same arguments have been made for students learning to read and write
in their second language. According to this perspective, the critical early
literacy-related and sociocultural experiences that ELLs have developed in
their L1 can be used to advantage during L2 literacy development. From a
pedagogical perspective, this could involve direct instruction in the L1 (as
in forms of bilingual education) or in English-only programs with some
kind of pedagogical recognition of the existing resources that ELLs have
already developed in the L1 (e.g., use of folktales that are familiar to ELLs).
In contrast, others have argued that promotion of ELLs’ L1 oral proficiency
detracts from their L2 development and especially the development of L2
literacy because it deprives these learners of valuable learning time in
the L2 (Porter, 1990; Rossell and Baker, 1996). This is sometimes referred
to as the “time-on-task” argument. This view assumes a sequential and
mono-linguistic relationship between oral language proficiency and literacy
development in a given language. Inherent in such a view is the notion
P1: JZZ
that L2 reading and writing development proceeds autonomously from

L1 proficiency. The studies reviewed in this section (see Table A.3.1 for a
summary of these studies) examine the effects of L1 oral proficiency on L2
literacy with respect to general aspects of L1 ability and with respect to
more specific aspects of L1 ability. We begin with the former.
The role that L1 oral proficiency plays in the development of L2 literacy
has been examined in global ways in terms of the extent of L1 use out-
side of the school setting. A number of these studies used national data
sources and multivariate research designs and found that use of a language
other than English at home had no or only a weak or indirect relationship
with literacy achievement in school (Buriel and Cardoza, 1998; Fernandez
and Nielsen, 1986; Kennedy and Park, 1994; Nielsen and Lerner, 1986).
There are a number of exceptions to this general trend, but even the effects
reported in these studies are circumscribed. More specifically, Buriel and
Cardoza (1998) compared three generations of Hispanic ELLs and found a
significant negative relationship between L1 oral proficiency and L2 read-
ing development in the third generation cohort, but found no relationship
between L1 proficiency and L2 reading development in the first and sec-
ond generation cohorts. Kennedy and Park (1994), comparing Hispanic
and Asian background ELLs, found that speaking a language other than
English at home had a negative relationship with standardized reading
test scores in English for ELLs with Asian backgrounds but not with other
measures of reading achievement. Moreover, no such effects were reported
for the Hispanic cohort. It is particularly noteworthy that, notwithstanding
these exceptions, all of the large-scale studies cited found that other factors,
such as socioeconomic status, sense of control, aspirations, and amount of
homework, were more significant predictors of reading achievement than
was L1 use outside of school. In other words, L1 language use was gen-
erally less predictive of subsequent L2 reading development than other
psychosocial factors. The link between L1 use and other factors outside of
school and L2 literacy development in school is discussed more compre-
hensively in the “Language of Instruction” section in the following chapter.
Generally speaking, studies that have examined the link between more
specific aspects of L1 oral ability or usage (e.g., emergent literacy, being
read to at home) and L2 reading and writing development in school report
“that early literacy experiences support subsequent literacy development,
regardless of language [emphasis added]; and time spent on literacy activ-
ity in the native language – whether it takes place at home or at school – is
not time lost with respect to English reading acquisition, at least through
middle school” (Reese et al., 2000, p. 633). More specifically, Reese et al.
(2000) found that family literacy practices, regardless of which language
they occurred in, and L1 emergent literacy were significant predictors of
L2 reading achievement in later grades. Following from this, a number of
studies found that ELLs can draw on L1 experiences and abilities to the
P1: JZZ
Literacy 69
benefit of their performance on L2 literacy tasks, especially when given

explicit opportunities to do so. Langer et al. (1990), using rigorous quali-
tative analyses, found that ELLs successfully made use of competencies in
their L1 to make sense of L2 reading tasks. For example, they found that
ELLs drew on words and ideas in their L1 especially when reading in the
L2 was difficult. Lanauze and Snow (1989) found that students who were
orally proficient in their L1 but not their L2 as well as students who were
proficient in both their L1 and L2 exhibited similar levels of complexity,
sophistication, and semantic content in their L2 writing. Lanauze and Snow
note that writing performance in the L2 can surpass oral proficiency in the
L2 in some cases. Accordingly, they go on, if proficiency is developed in the
L1 (Spanish), those skills can transfer easily to the L2. In further support of
the recruitment of the L1 in L2 reading, Saville-Troike (1984) reports (albeit
descriptively) that the majority of top achievers on measures of L2 reading
made use of their L1 during problem solving.
In summary, the findings outlined in this section suggest that, with some
exceptions, measures of general L1 language proficiency or usage outside
of school have not been found to relate consistently to the L2 literacy devel-
opment of ELLs in school. Viewed differently, use of the L1 does not seem to
detract from ELLs’ L2 literacy development. Furthermore, it would appear
that more specific measures of L1 oral proficiency or usage – and, in partic-
ular, those that are related to literacy – have a more significant and positive
developmental relationship with L2 literacy than do general oral language
proficiency measures. For example, ELLs with early L1 emergent literacy
experiences and skills appeared to be able to utilize these experiences in
the continued development of literacy abilities in the L2. In addition, ELLs
were able to draw on existing L1 oral abilities, either in the absence of sim-
ilar abilities in the L2 or in addition to similar abilities, in the service of L2
literacy tasks.
The role that L1 oral proficiency plays in L2 literacy should be consid-
ered in future research in more systematic ways, particularly with more
direct measures of L1 oral proficiency. In the majority of studies reviewed
here, L1 oral proficiency was assessed very generally, using self-report
or global indicators, or was simply assumed. Since much of the research
reviewed here suggests that certain levels and aspects of L1 oral proficiency
are related to L2 literacy development than others, more attention to the
precise nature of this relationship is needed if these relationships are to be
explicated clearly.
l2 oral proficiency and l2 literacy

Although a certain minimum level of general oral language proficiency
in L2 is undoubtedly necessary for L2 literacy development, the relation-
ship between L2 oral and L2 literacy development appears to be more
P1: JZZ
complex than the relationship between L1 oral language and L1 literacy.

As discussed in the previous section, L2 literacy often draws on knowl-
edge and experiences linked to the L1; thus, L2 oral proficiency is likely
to play a different role in the L2 literacy development of ELLs. In other
words, the contribution that L2 oral proficiency makes to L2 literacy devel-
opment in the case of ELLs may be composed of specific aspects of L2
oral proficiency that work in a complementary fashion with L1 oral profi-
ciency (Peregoy and Boyle, 1991). Research reviewed in this section (see
Table A.3.2) supports the notion that the development of L2 literacy can
proceed with limited L2 oral proficiency if students have sufficiently devel-
oped abilities in their L1 (e.g., Lanauze & Snow, 1989; Reese et al., 2000).
In such cases, it appears that L1 oral proficiency and emergent literacy in
the L1 can fill gaps in L2 oral proficiency as it develops. This does not
mean that L2 oral proficiency does not contribute to L2 literacy develop-
ment since, as Reese et al. (2000) have noted, ELLs who begin school with
well-developed L2 oral abilities achieve greater success in English reading
than children with less well-developed L2 oral language abilities. How-
ever, the findings from these studies underline the important contribution
that L1 abilities can make to L2 literacy development when dealing with
students with limited L2 oral proficiency. Furthermore, a consideration of
the differential roles that L1 and L2 oral proficiency might play in relation
to L2 literacy development could help to define more clearly a number of
important constructs that are often used when investigating these issues;
specifically, the constructs of developmental interdependence, common
underlying proficiency, and the thresholds of oral proficiency necessary to
promote L2 literacy development discussed earlier.
Lindholm and Aclan (1991) sought to identify whether or not there
is a threshold level of bilingual proficiency that results in enhanced lev-
els of L2 reading achievement, as proposed by the threshold hypothesis
(Cummins, 1991). More specifically, they examined the relationship among
high, medium, and low levels of bilingual proficiency and English L2 read-
ing achievement among Grades 1 to 4 elementary school ELLs. Since the
students’ levels of bilingual proficiency varied primarily with respect to
level of L2 proficiency (with L1 oral proficiency assumed), their study per-
mits us to examine the link between L2 oral proficiency and L2 literacy.
The authors report that although there was a trend for scores on English
reading measures to increase in accordance with increasing levels of bilin-
gual proficiency, there was no significant difference between proficiency
groups on reading measures in Grades 1 and 2. However, by Grade 3, the
same year in which English reading instruction was introduced, differential
effects of bilingual proficiency were evident with high levels of bilingual
proficiency being significantly related to high levels of L2 reading ability.
By implication, these results suggest that high levels of L2 oral proficiency
P1: JZZ
Literacy 71
can enhance L2 literacy development to a significant extent. In support of

the threshold hypothesis, they also found that only the highly proficient
bilingual students reached grade-level norms in English by Grade 4. In
concluding, Lindholm and Aclan (1991) emphasize the need to evaluate
student achievement in bilingual education programs from a long-term
developmental perspective in order to determine the true effects of bilin-
gualism on L2 literacy development (see also Cummins, 1992).
The remaining studies reviewed in this section focused on discrete
aspects of L2 oral proficiency, to identify those specific features of L2 oral
proficiency that contribute significantly to L2 literacy development. Stud-
ies that have addressed this issue have identified a differential relation-
ship between L2 literacy achievement, on the one hand, and specific facets
of L2 oral language proficiency on the other hand, with L2 oral abilities
that are linked to academic tasks being more highly related to L2 liter-
acy development than general L2 oral proficiency. Saville-Troike (1984)
found that diversity of L2 vocabulary was significantly related to reading
achievement, whereas general oral proficiency and verbosity were not. In
a multiple-case study, Peregoy (1989) compared the L2 reading abilities of
high, intermediate, and low L2 oral proficiency groups. She found a gen-
eral correspondence between levels of L2 oral proficiency and L2 reading
comprehension, and evidence for differential effects of specific compo-
nents of oral proficiency at different proficiency levels. Lack of vocabulary
knowledge resulted in reading miscomprehension at all levels, but it was
particularly detrimental for low-level students, where lack of syntactic
knowledge also impeded reading comprehension. Perez (1981) found that
direct instruction in aspects of L2 oral competence specifically related to
literacy (e.g., multiple word meanings, sentence patterns) resulted in sig-
nificant improvements to the L2 reading scores of third-grade ELLs. In a
study by Royer and Carlo (1991), L2 listening comprehension, as measured
by performance on a sentence verification task, was a significant predictor
of L2 reading performance, second in importance only to L1 or L2 reading
scores.
Goldstein, Harris, and Klein (1993) and Peregoy and Boyle (1991) both
examined the relationship between L2 reading comprehension and L2 oral
proficiency as measured by knowledge of surface structure elements (e.g.,
grammatical complexity) versus deep structure elements (e.g., informa-
tiveness of responses). Goldstein, Harris, and Klein (1993) used a story-
retelling task as a measure of L2 oral proficiency. The students’ oral
retellings were scored in two different manners, first for surface structure
features and understanding and second for underlying story structure and
in-depth understanding. They found that the results from the deep struc-
ture analyses were more highly related to L2 reading comprehension than
were the results for the surface structure features. Peregoy and Boyle (1991)
P1: JZZ
compared high, medium, and low level reading proficiency groups on four
oral proficiency measures, two that reflected relatively surface level lin-
guistic abilities and two that reflected deeper cognitive-linguistic abilities.
Their results provide some evidence for the differential effects of deep ver-
sus surface structure features insofar as the intermediate and high groups
differed significantly on two of the four measures, including the deep struc-
ture feature of “informativeness.”
These studies considered together provide evidence that certain features
of L2 oral proficiency are more directly related to L2 literacy than others.
However, since the specific aspects that have been examined are diverse,
it is difficult to draw definitive conclusions about which specific features
are consistently related to improvements in L2 reading and writing perfor-
mance. Clearly, more research is needed to clearly identify those aspects of
L2 oral proficiency that contribute more directly and reliably to L2 literacy
development.
In sum, findings from research in this and the preceding section on the
link between L1 and L2 oral proficiency and L2 literacy development pro-
vide evidence for both crosslinguistic and crossmodal effects. It appears
that aspects of both L1 and L2 oral proficiency are linked to L2 literacy
development and that the relationship between oral and literacy develop-
ment in ELLs is more specific and complex than might have previously
been thought. The link between L2 oral proficiency and L2 literacy that is
revealed by extant research points to a nuanced role for L2 oral language
development, with academic- and literacy-related L2 oral proficiency being
more important than general communicative competence in the L2. At the
same time, the contribution of specific L2 oral language abilities to L2
literacy development needs to be considered with reference to the linguis-
tic knowledge and real-world experiences that ELLs acquire through the
medium of their L1. That is to say, it would appear that L1 oral language
experiences and knowledge are critical developmental factors in ELLs’
L2 literacy development, and that L2 oral proficiency may contribute in
a complementary and specific manner. Furthermore, the relationship of
oral proficiency in both the L1 and L2 needs to be considered more specif-
ically in terms of how they might contribute to a common underlying
proficiency.
In the following section, we review research that focuses on discrete
aspects of L2 literacy development, often referred to as component skills
or abilities, such as phonological awareness and vocabulary development.
Because these components are, arguably, more easily definable and mea-
surable than other, more complex aspects of reading and writing develop-
ment, they have yielded relatively clear results and might serve as a basis
for conducting further research on aspects of L2 literacy development that
are more complex in nature.
P1: JZZ
Literacy 73
components of literacy development

The studies reviewed in this section (see Table A.3.3) are diverse in their
focus and approaches but are considered together because all look at spe-
cific components of reading and writing. At issue is the extent to which
L2 literacy development is influenced by common underlying language-
related abilities that apply to virtually any language, as in studies dealing
with phonological awareness or, conversely, by language-specific abilities
that emanate from the L1 or the L2, as in studies dealing with spelling or
cognate vocabulary.
Phonological Awareness
Research on L1 reading has established that phonological awareness is a
significant correlate of successful beginning reading development (Adams,
1990). Phonological awareness is the ability to analyze and manipulate the
sounds that make up oral and written language and can be assessed using
a variety of tasks: rhyme detection – what two words sound the same at the
end: pat, pan, cat?; phoneme deletion – what sound is left if the sound “k” is
removed from the word “cat”; blending – what word do you get if the sound
“s” is added to the word “top.” The causal relationship between read-
ing and phonological awareness has been shown to be bidirectional, with
certain aspects of phonological awareness playing a fundamental role in
facilitating early reading acquisition, while reading acquisition itself facil-
itates the emergence of yet other, more sophisticated aspects of phonolog-
ical awareness (Adams, 1990; Snow, Burns, and Griffin, 1998). The causal
role of phonological awareness in reading acquisition is also supported by
intervention studies that show that children with difficulty learning to read
exhibit statistically significant gains in reading ability following training
in phonological awareness (Torgesen et al., 2001) and also by research that
shows that poor and good L1 readers differ significantly from one another
on tasks that tap phonological awareness, suggesting that phonological
awareness is a decisive factor (Wagner and Torgesen, 1987).
The research reviewed in this section examines phonological awareness
in L1 and L2 and its relationship to L2 reading. A critical question at the
heart of this research is whether phonological awareness and its relation-
ship to reading acquisition is tied to a particular language or whether it
is a meta- or common underlying linguistic ability that has crosslinguistic
repercussions, as noted by Durgunoglu, Nagy, and Hancin-Bhatt (1993,
p. 454): “The ability to hear small components of spoken language may be
highly correlated between languages.” The corpus of research reviewed
here is small since our literature search was limited to studies that exam-
ined the link between phonological awareness and reading directly; that is
P1: JZZ
to say, the study had to have measures of reading to be included. Although

few in number and diverse in focus, the studies reviewed here all point
toward the same general conclusion; namely, that phonological awareness
is a common underlying ability that is linked to oral language develop-
ment and is shared crosslinguistically – phonological awareness in one
language (e.g., L1) supports phonological awareness in an additional lan-
guage (e.g., L2) and, in turn, reading acquisition in that language. The
results from instructional studies also suggest that phonological aware-
ness in the L2 can be developed through direct intervention, even if L2
oral development is itself somewhat limited, adding further evidence
that phonological awareness is a metalinguistic or common underlying
proficiency.
That L2 phonological awareness is significantly related to L2 reading
development, as L1 phonological awareness is linked to L1 reading devel-
opment, is evidenced in research by Carlisle et al. (1999). They found that
English L2 phonological awareness contributed to English L2 reading com-
prehension. Durgunoglu, Nagy, and Hancin-Bhatt (1993) point to phono-
logical awareness as fundamentally crosslinguistic in nature based on their
finding that Spanish L1 phonological awareness was a significant predic-
tor of English L2 word recognition. The crosslinguistic interdependence
of phonological awareness is supported further in an interesting study
by Roberts and Corbett (1997) that showed that instruction in English
L2 phonological awareness significantly improved Hmong L1 phonologi-
cal awareness. Evidence for the trainability of L2 phonological awareness
comes from Roberts and Corbett (1997) and Terrasi (2000), who found that
direct instruction in phonological awareness in L2 English significantly
enhanced phonological awareness in that language. That phonological
awareness can be promoted independently of general L2 oral proficiency
is supported by Durgunoglu at al. (1993), who found that L1 phonological
awareness was a more significant predictor of L2 word reading ability than
were either L1 or L2 oral proficiency.
Orthographic Knowledge
While the findings from studies of phonological awareness argue for
crosslinguistic influences that are common in learning any language, stud-
ies that have examined sound-letter correspondences and spelling report
evidence for both language-specific and common developmental influ-
ences. Thus, on the one hand, it appears that L2 spelling is subject to con-
trastive L1–L2 effects in line with a contrastive analysis perspective – that
is to say, differences in sound-letter correspondence in the L1 and L2 can
result in negative transfer from the L1. On the other hand, ELLs’ English
spelling patterns have been shown to reflect developmental processes that
are also exhibited by native English speakers.
P1: JZZ
Literacy 75
Evidence of negative transfer in spelling comes from studies by Fashola

et al. (1996) and Zutell and Allen (1988), who found that Hispanic ELLs
erroneously applied Spanish L1 phonological and orthographic rules when
asked to write selected words with contrastive English/Spanish spelling
patterns. In a descriptive analysis of writing samples, Cronnell (1985) also
identified L1 influences in L2 errors. In contrast, Tompkins, Abramson,
and Pritchard (1999) failed to find such negative transfer when they exam-
ined naturally occurring spelling errors in the writing journals of ELLs
from different language backgrounds and English L1 children; the authors
suggest that the ELLs may have avoided using words with contrastive
patterns in order to avoid errors. The only errors differentiating the ELLs
and English L1 students in the Tompkins et al. study were those involving
inflectional endings, a finding also reported by Cronnell (1985). The stu-
dents exhibited largely developmental patterns in their English spelling,
patterns that were also exhibited by native English speakers. Such target-
like error patterns argue for developmental language learning processes
that characterize both native speakers and L2 language learners of the same
language.
Research by Hsia (1992) that used both phonological and spelling mea-
sures to examine L1 transfer effects on L2 development suggests that such
effects may be more likely in the early or beginning stages of develop-
ment when learners lack knowledge of more appropriate, target-like fea-
tures of the new language. More specifically, Hsia examined the influ-
ence of Chinese-background ELLs’ knowledge of L1 Mandarin syllable
segmentation patterns on their phoneme and syllable segmentation abili-
ties in English and found that, although there was an initial Mandarin L1
effect, English native-like phonological constraints were subsequently and
quickly acquired.
Cognate Vocabulary
Research on ELLs’ recognition and use of cognate relationships between
L1 and L2 vocabulary has shown that ELLs can make use of L1 vocabulary
knowledge to determine the meaning of cognate vocabulary in L2 text. All
of the research on this issue has examined ELLs of Hispanic background.
More specifically, Nagy et al. (1993) and Jimenez, Garcia, and Pearson (1996)
found that more successful L2 readers were better able than less successful
L2 readers to explicitly recognize Spanish-English cognates and to make
use of their knowledge of cognates during reading. These researchers,
as well as Hancin-Bhatt and Nagy (1994), also found that the ability to
translate cognates from L2 to L1 was linked to individual students’ pref-
erence to speak Spanish and their level of bilingualism and, in particular,
their knowledge of Spanish vocabulary, arguing, once again, that ELLs’
L1 need not be a distracting but rather a facilitating factor in L2 literacy
P1: JZZ
development. Finally, Nagy et al. (1993) and Hancin-Bhatt and Nagy (1994)
have found that Spanish L1 ELLs are better able to make use of spelling than
morphological similarities to recognize cognates, although use of morpho-
logical similarities increased with grade level. Thus, instruction in specific
morphological similarities between cognates might contribute to the L2
literacy development of ELLs by enhancing their knowledge of these oth-
erwise underused cognate relationships.
In sum, findings from research on specific components of reading and
writing support the conclusion that L2 literacy development can be influ-
enced by both common or metalinguistic abilities as well as by features of
language specific to the L1 or L2. Research focusing on phonological aware-
ness provides clear evidence that such awareness appears to emanate from
a common underlying ability that can be developed either through the L1
or the L2 and is manifested in both L1 and L2 literacy development in vir-
tually the same way. However, more crosslinguistic studies of metaphono-
logical awareness are needed to ascertain to what extent and in what ways
this is true. Research also indicates that such metalinguistic abilities can
be developed autonomously in the L2, even when learners have limited
proficiency in the L2.
Research concerned with ELLs’ orthographic development found evi-
dence for influences from the L1 as well as from the L2, in the latter case
leading to developmental patterns that are similar to those of native speak-
ers of the L2. Research that has investigated cognate relationships between
vocabularies in the L1 and the L2 provides a clear example of how ELLs
can draw on knowledge that is specific to the L1 in developing vocabulary
in the L2.
Research focusing on orthographic patterns and cognate relationships
between languages both suggest that ELLs can benefit from direct instruc-
tion about systematic functional and structural differences and similarities
between languages and such instruction enhances crosslinguistic facili-
tation. Arguably, the use of L1 language-specific knowledge or abilities
during L2 literacy tasks may serve to fill gaps in the learners’ competence
when they have not yet acquired target-appropriate knowledge of the L2.
Learning patterns that echo those of native-speaking readers and writers
seem to emerge as L2 learners advance in their L2 literacy development,
as is to be expected. Clearly, transfer of orthographic and cognate vocab-
ulary knowledge is more likely in languages that are typologically similar
(Spanish and English) than for typologically different languages (e.g.,
English and Chinese). Much more research is needed to explore cross-
language relationships in languages that are not alphabetic and that are
otherwise typologically different from English.
Further research in these areas, especially with different language pairs,
is needed to further our knowledge of the precise nature of putative
P1: JZZ
Literacy 77
common underlying abilities, as well as to determine how systematic rela-

tionships between the L1 and the L2 can be exploited by ELLs in their L2
literacy development.
l1 literacy and l2 literacy development

Research on the effects of L1 literacy on L2 literacy development is the final
issue in our survey of crosslinguistic/crossmodal relationships. Although
the “time on task” view of L2 development might oppose promotion of
L1 literacy on the grounds that it reduces the time ELLs have to devote
to L2 literacy development and others argue that it is a source of inter-
ference or confusion, research such as that by Nguyen and Shin (2001)
supports the view that competence in L1 literacy does not retard L2 lit-
eracy development. Much of the evidence concerning the effects of L1
literacy development on L2 literacy development comes from research on
program comparisons, initial language of instruction, and various instruc-
tional strategies – all of which is reviewed in other chapters. These studies,
reviewed elsewhere in this volume, examined this issue in relatively gen-
eral terms by comparing students’ general levels of reading and writing
achievement in both languages.
What remains to be discussed in this section are more specific develop-
mental relationships between the two literacies; that is, the specific ways
in which L2 literacy develops in bilingual contexts. This has been exam-
ined by examining specific aspects of literacy and specific types of learners
(e.g., successful and unsuccessful ELL readers/writers). In this section,
the concept of a common underlying proficiency (as proposed by Cum-
mins) is reconceptualized as a common underlying reservoir of literacy
abilities. The concept of a reservoir suggests that ELLs develop various
literacy abilities that are then available to them when reading or writing in
either language. Furthermore, ELLs who remain proficient in their L1 also
develop abilities specific to their bilingual state (e.g., use of cognates); thus,
the concept of a reservoir becomes a bilingual reservoir. As the following
section sugggests, differences in literacy performance may be due to dif-
ferent ways in which ELLs develop and draw on this common underlying
reservoir of literacy abilities. A summary of the research included in this
section of our synthesis is provided in Table A.3.4.
Effects of L1 Literacy on L2 Literacy Development

A number of studies mentioned previously with regard to the relationship
between oral proficiency and L2 literacy also explored the effect of L1 liter-
acy on the development of L2 literacy (e.g., Royer and Carlo, 1991; Langer
et al., 1990; Reese et al., 2000). These studies found that the relationship
P1: JZZ
between literacy in the L1 and the L2 is at least as significant as, if not

more significant than, that between L2 oral development and L2 literacy.
These findings, in turn, argue that developing literacy in the L1 does not
detract from literacy development in the L2 but rather supports it. To be
more specific, Reese et al. (2000), discussed previously, found that ELLs
who were identified as the best L1 readers were deemed able to transi-
tion to English reading instruction earlier than other students and that
early L1 reading abilities were a significant predictor of English reading
abilities assessed eight years later. Additional evidence in support of the
additive effects of L1 literacy development comes from Collier (1987) who,
in a cross-sectional study, examined the link between length-of-residence
and age-of-arrival on ELLs’ English reading achievement. She found that
late elementary grade ELLs with at least two years of L1 reading instruc-
tion reached grade-level equivalence in English reading more rapidly than
those with little or no schooling in the L1. However, older ELLs (who
arguably face relatively cognitively demanding L2 academic tasks) did
not achieve grade-level equivalence as quickly as younger ELL students
despite the fact that the former had had more years of L1 literacy devel-
opment, suggesting that the issue is complicated by the nature and level
of the reading tasks required of the learner. Royer and Carlo (1991) found
that the L1 reading abilities of ELLs in Grade 5 were the best predictor of
their L2 reading achievement in Grade 6, thereby providing corroborative
evidence for the supportive effect of the L1. These findings suggest that
L1 literacy needs to develop to a certain level if it is to benefit L2 literacy
development.
A number of studies that have looked at the acquisition and use of
specific literacy abilities across languages corroborate this general rela-
tionship. Buriel and Cardoza (1998), Lanauze and Snow (1989), Langer et
al. (1990), Jimenez et al. (1996), and Jimenez (2000) have all found evidence
for specific parallel abilities across languages. For example, in a study of L2
writing among Grades 4 and 5 Hispanic ELLs, Lanauze and Snow (1989)
found that ELLs exhibited similar profiles with respect to the complexity,
sophistication, and semantic content of their writing in both their L1 and
L2; this was evident even for students who were not orally proficient in
their L2. These findings suggest that ELLs are able to apply abilities devel-
oped in their L1 to L2 literacy tasks. This pattern is further illustrated when
the literacy profiles of successful and less successful readers are compared.
Langer et al. (1990), Jimenez et al. (1996), and Jimenez (2000) found that
ELLs who were successful readers were successful in both languages, and
ELLs who were unsuccessful readers were unsuccessful in both languages.
These studies all uphold the notion that successful literacy development in
both languages appears to be supported by a common underlying reservoir
of literacy abilities and that L1 literacy can contribute to the development
of this reservoir of abilities.
P1: JZZ
Literacy 79
L1 and L2 Literacy Strategies

Research that has examined the strategies used by ELLs during L2 literacy
tasks provides further insight into the nature of the additive relationship
between L1 and L2 literacy. Research in this corpus has examined this issue
in two ways: by comparing the strategies used by ELLs during both L1 and
L2 literacy tasks; and by comparing the strategies used by ELLs during L2
English reading tasks with those used by native English speakers.
Research that has compared the strategies used by ELLs during L1
and L2 literacy tasks has found that successful and “unsuccessful” ELL
readers/writers employ different strategies (Calero-Breckheimer and
Goetz, 1993; Jimenez, 2000; Jimenez et al., 1996; Langer et al., 1990,
Miramontes, 1987). More specifically, but perhaps not surprising, success-
ful ELL readers/writers employ a number of effective strategies, such as
using context and inferencing, monitoring comprehension, and invoking
prior knowledge, whereas unsuccessful ELL readers employ a variety of
ineffective or less sophisticated strategies (Padron and Waxman, 1988).
They fail to draw or adjust inferences; they often invoke irrelevant prior
knowledge; and they view completion as more important than compre-
hension (Jimenez et al., 1996). Of perhaps more interest, this research also
found that successful readers/writers demonstrate use of the same strate-
gies during both L1 and L2 literacy tasks, and they view reading in the L1
and L2 as similar activities or processes with language-specific differences.
Jimenez et al. (1996) reported that successful ELL readers/writers were able
to deploy a variety of effective “bilingual” strategies, such as searching for
cognates, judicious translation, or use of prior knowledge developed in the
L1. In contrast, the less successful ELLs viewed reading in the L1 and the
L2 as separate abilities and saw the L1 as a source of confusion. That the
unsuccessful ELL readers/writers viewed L1 and L2 reading in these ways
suggests that they had not developed an understanding of the common-
alities in L1 and L2 literacy and, as a result, were unable to draw on sim-
ilarities and connections between their two languages in the service of L2
reading and writing. Jimenez (2000) suggests that unsuccessful ELL read-
ers may need opportunities to learn about similarities between the writing
systems of their two languages and to become more aware of bilingual
strategies that would encourage them to draw on knowledge resources in
the L1 to enhance their literacy abilities in the L2 (see also Langer et al.,
1990).
At the same time, research that has compared L2 (i.e., ELL) with L1
English readers/writers has found that their strategies differ. More specif-
ically, L1 English readers have been shown to use significantly more and
different strategies in general than ELL readers (Padron, Knight, and Wax-
man, 1986; Knight, Padron, and Waxman, 1985). Bean, Levine, and Graham
(1982) and Miramontes (1987, 1990) found that ELLs pay closer attention to
P1: JZZ
textual features than L1 English readers. For example, Miramontes (1987,

1990) found that good Spanish readers paid significantly more attention
to graphic representation and grammatical structure in both L1 and L2
reading than good English readers. Although these researchers report no
apparent loss in comprehension by ELLs, they suggest that the strategies
used by ELLs in their English reading are inappropriate because they are
not the same as those employed by successful L1 English readers. However,
the reading performance of the ELLs as reported in this study does not back
up this claim. Rather, the pattern of strategies employed by successful ELL
readers and writers may be more appropriately construed as an equally
effective but different path to literacy development in comparison to that
exhibited by L1 readers and writers.
An explanation of the differences between successful ELL and L1 read-
ers can be offered in terms of the former’s having access to a bilingual
reservoir of literacy abilities and strategies in contrast to the latter’s mono-
lingual pool of resources. Langer et al. (1990) and Jimenez et al. (1996)
add support to this possibility by providing evidence that successful ELL
readers maximize what they know by using their L1 to translate, elaborate,
and hypothesize when making sense of English text. Edelsky and Jilbert
(1985) have made a similar claim: “children’s bilingualism increases their
options for making meaning” (p. 69). Such a notion sees reading in an L2
as part of a larger, bilingual process. Such a process is also supported by the
research discussed earlier with respect to L1 spelling patterns and cognate
vocabulary (see also Nagy, McClure, and Mir, 1997, for evidence concern-
ing L1 syntactic influences on determining unfamiliar word meanings in
L2 reading). It follows that attempts to get ELLs to adopt strategies that are
similar to those of monolingual English readers may be misguided because
they fail to acknowledge and draw on the full capacities of bilingual learn-
ers, which necessarily encompass contributions and knowledge from two
languages.
Other Issues: Text Types and Genre

We finish this chapter with a final issue that deserves consideration but
has received scant empirical attention; namely, ELLs’ need to develop pro-
ficiency in different types and genres of literacy if they are to achieve their
full literacy potential. Jimenez et al. (1996) and Langer et al. (1990) note that
some ELLs have difficulty with academic or cognitively demanding types
of texts (e.g., they find reports more difficult to read and understand than
stories). Jimenez et al. (1996) also note, in comparing successful L1 English
readers with successful ELL readers in English, that the two groups dif-
fered qualitatively in terms of their concern for detail and the types and
level of sophistication of the connections they made during literacy tasks.
Bermudez and Prater (1994) suggest the need to provide opportunities
P1: JZZ
Literacy 81
for ELLs to develop more sophisticated expertise in the use of persuasive

discourse, while Langer et al. (1990) observe that ELLs who are transi-
tioning to English literacy have difficulty interpreting decontextualized
reading comprehension questions, a finding also reported by Field (1996)
in a descriptive study of ELLs who were transitioning to English literacy.
Two studies that differ significantly in their focus, and due to the case-
study nature need to be interpreted with caution, also support the notion
that ELL readers/writers can develop social and critical aspects of L2 lit-
eracy in ways similar to that of English L1 readers/writers (Samway, 1993;
Urzua, 1987). Bringing these broader literacy issues back to a consideration
of the construct of common underlying proficiencies, as Jimenez (2000) and
Galindo (1993) suggest, ELLs need literacy-development experiences that
are connected to their bilingual abilities and bicultural status.
In summary, research that has sought to define the relationship between
L1 literacy and L2 literacy has found that L1 literacy does not detract from
L2 literacy development but rather contributes to and supports its develop-
ment. In effect, ELLs with successful L1 literacy experience progress more
quickly and successfully in their L2 literacy development than ELLs with
no prior L1 literacy. Research also provides evidence for parallel abilities
across languages, thereby supporting the construct of a common under-
lying reservoir of literacy abilities. In brief, the evidence reviewed in this
section indicates that there can be additive developmental effects of L1
literacy development on L2 literacy development. This does not mean that
L2 development is unimportant because, as Reese et al. found, ELLs with
pre-school literacy experiences in either the L2 or the L1 are advantaged
with respect to later reading acquisition. This issue is discussed further in
Chapter 4.
Research that has examined the strategies employed by ELLs in L1 and
L2 literacy tasks provides further insight into the processes of L2 reading
and writing. Studies show that successful ELL readers and writers use
similar strategies in both languages, whereas less successful ELL readers
and writers do not, apparently not capitalizing on the commonalities of
literacy across languages. Furthermore, successful ELLs also make use of
effective strategies not available to successful monolinguals; strategies that
draw on knowledge of and relationships between the L1 and the L2.
Finally, studies that have focused on the context and content of liter-
acy activities suggest that more attention needs to be given to developing
ELLs’ abilities with respect to deeper and more cognitively demanding
aspects of literacy. Research has shown that certain text types, such as fac-
tual reports as opposed to narratives, pose more difficulty for ELLs, as do
more decontextualized literacy tasks. Researchers suggest that develop-
ment and success in these more demanding literacy tasks can be facilitated
by drawing on ELLs’ sociocultural knowledge, including their L1 as well
as L2 experiences.
P1: JZZ
conclusions
The various L1 and L2 as well as oral and written proficiencies discussed
in this chapter contribute in different yet complementary ways to L2 liter-
acy development. These abilities appear to contribute to the development
of a common bilingual reservoir that serves both L1 and L2 literacy and
create an awareness of systematic relationships between languages, allow-
ing ELLs to draw on existing L1 knowledge in the service of L2 literacy.
Furthermore, it appears L2 literacy is, in a sense, more than the sum of
its parts, as ELLs appear to have unique abilities that result from their
bilingual status.
Research that has focused on the relationship between L2 literacy and
oral language proficiency in the L1 and L2 reveals a relationship between
oral and written language in ELLs that is specific and complex. In partic-
ular, research that has examined the influence of L1 oral proficiency on L2
literacy found that not only does it not detract from L2 literacy develop-
ment, but also that specific aspects of L1 oral language proficiency, such as
L1 emergent literacy, are more influential in L2 literacy development than
general aspects of L2 oral proficiency. It also appears that ELLs make use of
L1 oral proficiency to draw on prior knowledge and experience, either in
the absence of or in addition to similar levels of L2 oral proficiency, in the
service of L2 literacy tasks. Findings from research concerned with the rela-
tionship between L2 oral proficiency and L2 literacy development suggest
that a certain level of L2 oral proficiency needs to be attained for a sig-
nificant relationship to be evident. Furthermore, as with L1 oral language
proficiency, specific literacy-related aspects of L2 oral proficiency, such as
diversity of vocabulary and in-depth text understanding, appear to be more
highly related to L2 literacy abilities than do more general or surface-level
L2 oral abilities. Moreover, it appears that L2 literacy development can
proceed to some extent even with limited L2 oral proficiency, provided
that consideration is given to linguistic and prior experiential knowledge
that ELLs have already acquired through the medium of their L1. If future
research supports this conclusion, it would follow that instructional con-
sideration of aspects of both L1 and L2 oral language could optimize L2
literacy development, arguably beyond what can be achieved through the
L2 alone. In sum, L1 and L2 oral abilities can contribute to L2 literacy
development in a complementary fashion.
Results from research that has examined specific components of reading
and writing further define L2 literacy development to include a complex
set of influences, including common underlying proficiencies, influences
from the L1, the learners’ knowledge of relationships between their lan-
guages, and typical developmental processes linked to the L2. Research
on phonological awareness suggests that such awareness is a common
underlying ability that, once acquired, is manifest in both L1 and L2
P1: JZZ
Literacy 83
literacy development. The findings reviewed here suggest that phonolog-

ical awareness can be developed through the L1 and applied to the L2
or developed through the L2, even if ELLs have limited L2 proficiency.
The influence of the L1 is evident from research that has looked at the
development of L2 spelling. Studies using word lists that contrast spelling
patterns between the L1 and the L2 show L1 influence or “negative transfer
in ELLs” spelling errors. On the other hand, studies that examined spelling
errors spontaneously produced by ELLs found that other types of errors
correspond to developmental patterns specific to the L2, similar to those
made by native speakers. Research that has looked at the effect of cognate
relationships between the L1 and L2 on L2 literacy development provides
specific evidence of how ELLs can utilize knowledge of the L1 in acquiring
vocabulary in the L2. However, this same research indicates that knowl-
edge of specific orthographic and morphological correspondences can be
enhanced, suggesting that there is a potential for crosslinguistic facilitation
that is underutilized in L2 literacy development.
Perhaps the most direct crosslinguistic relationship discovered in this
review is that between L1 literacy and L2 literacy. Research on this relation-
ship finds that L1 literacy supports L2 literacy development. ELLs with
initial L1 literacy experiences, such as emergent and family literacy, as
well as those with well-developed L1 literacy experiences, progress more
quickly and successfully in L2 literacy development. Research findings
reviewed here also provide evidence for parallel abilities across languages,
supporting the common underlying proficiency, or reservoir of abilities,
construct. These parallel abilities are evidenced quite consistently in stud-
ies that focus on the strategies used by ELLs in L1 and L2 literacy tasks,
where ELLs who are successful in L2 literacy tasks use similar strategies in
both their L1 and their L2, viewing literacy in either language as a similar
event. Less successful ELL readers and writers use different and less effec-
tive strategies and see L1 and L2 literacy tasks as different. Furthermore,
ELLs appear to utilize different yet effective strategies in L2 literacy tasks
in comparison to monolinguals, strategies that appear to stem from their
bilingualism. The research reviewed in this section supports an additive
effect of L1 literacy on L2 literacy development and support the construct
of a common bilingual reservoir. It also provides the basis and impetus for
future research to investigate ELLs’ literacy abilities in both languages in
terms of shared and effective strategies, as well as their unique bilingual
strategies.
The final set of studies reviewed in this chapter calls attention to an
additional set of issues related to the content and types of literacy tasks
that ELLs confront in school. This research suggests that ELLs need more
exposure to and instruction relevant to complex genres of literacy.
When considered together, the crosslinguistic and crossmodal influ-
ences on L2 literacy development that have been reviewed in this chapter
P1: JZZ
form a complex yet coherent picture. At the same time, it is important to

note that the picture is at best preliminary and considerably more research
in most domains is required to draw stable and definitive conclusions.
References
Adams, M. 1990. Beginning to read. Cambridge: MIT Press.
Bean, T., Levine, M. G., & Graham, R. C. 1982. Beginning ESL readers’ attention to
the graphemic features of print. Reading Improvement 18(4), 346–9.
Bermudez, A. B., and Prater, D. L. 1994. Examining the effects of gender and sec-
ond language proficiency on Hispanic writers’ persuasive discourse. Bilingual
Research Journal 18(3–4), 47–62.
Buriel, R., and Cardoza, D. 1998. Sociocultural correlates of achievement among
three generations of Mexican-American high school seniors. American Educa-
tional Research Journal 25(2), 177–92.
Calero-Breckheimer, A., and Goetz, E. 1993. Reading strategies of biliterate children
for English and Spanish texts. Reading Psychology: An International Quarterly 14,
177–204.
Carlisle, J., Beeman, M., Davis, L., and Sparim, G. 1999. Relationship of metalin-
guistic capabilities and reading achievement for children who are becoming
bilingual. Applied Psycholinguistics 20(4), 459–78.
Collier, V. 1987. Age and rate of acquisition of second language for academic
Cronnell, B. 1985. Language influences in the English writing of third- and sixth-
grade Mexican-American students. Journal of Educational Research 78, 168–73.
Cummins, J. 1979. Linguistic interdependence and the educational development
of bilingual children. Review of Educational Research 49, 222–51.
1981. The role of primary language development in promoting educational
success for language minority students. In Schooling and Language Minority
Students: A Theoretical Framework (pp. 3–49). Los Angeles: National Dissemi-
nation and Assessment Center.
1991. Interdependence of first- and second-language proficiency in bilingual
children. In E. Bialystok (ed.), Language Processing in Bilingual Children (pp. 70–
89). Cambridge, UK: Cambridge University Press.
1992. Empowerment through biliteracy. In J. V. Tinajero and A. F. Ada (eds.), The
power of two languages: Literacy and biliteracy for Spanish-speaking students. New
York: McGraw-Hill.
1997. Cultural and linguistic diversity in education: A mainstream issue?
Educational Review 49, 99–107.
2000. Language, power and pedagogy: Bilingual children in the crossfire. Clevedon,
UK: Multilingual Matters.
Durgunoglu, A., Nagy, W., and Hancin-Bhatt, B. 1993. Cross-language transfer of
phonological awareness. Journal of Educational Psychology 85(3), 453–65.
Edelsky, C., and Jilbert, K. 1985. Bilingual children and writing: Lessons for all of
us. Volta Review 87(5), 57–72.
Ellis, R. 1986. Understanding second language acquisition. Oxford: Oxford University
Press.
P1: JZZ
Literacy 85
Fashola, O., Drum, P., Mayer, R., and Kang, S. 1996. A cognitive theory of ortho-
graphic transitioning: Predictable errors in how Spanish speaking children
spell English words. American Educational Research Journal 33(4), 825–43.
Fernandez, R. M., and Nielsen, F. 1986. Bilingualism and Hispanic scholastic
achievement: Some baseline results. Social Science Research 15(1), 43–70.
Field, M. 1996. Pragmatic issues related to reading comprehension questions: A
case study from a Latino bilingual classroom. Issues in Applied Linguistics 7(2),
209–24.
Galindo, R. 1993. The influence of peer culture on Mexican-origin bilingual chil-
dren’s interpretations of a literacy event. The Bilingual Research Journal 17(3–4),
71–99.
Goldstein, B., Harris, K., and Klein, M. 1993. Assessment of oral storytelling abil-
ities of Latino junior high school students with learning handicaps. Journal of
Learning Disabilities 26(2), 138–43.
Hancin-Bhatt, B., and Nagy, W. 1994. Lexical transfer and second language mor-
phological development. Applied Psycholinguistics 15(3), 289–310.
Heath, S. 1983. Ways with words. Cambridge, UK: Cambridge University Press.
Hsia, S. 1992. Developmental knowledge of inter- and intraword boundaries:
Evidence from American and Mandarin Chinese speaking beginning readers.
Applied Psycholinguistics 13(3), 341–72.
Hudelson, S. 1994. Literacy development of second language children. In F. Genesee
(ed.), Educating second language children: The whole child, the whole curriculum, the
whole community (pp. 129–58). Cambridge, UK: Cambridge University Press.
Jimenez, R. T. 2000. Literacy and the identity development of Latina/o students.
American Educational Research Journal 37(4), 971–1000.
Jimenez, R., Garcia, G. E., and Pearson, P. D. 1996. The reading strategies of bilin-
gual Latina/o students who are successful English readers: Opportunities and
obstacles. Reading Research Quarterly 31(1), 90–112.
Kennedy, E., and Park, H. 1994. Home language as a predictor of academic
achievement: A comparative study of Mexican- and Asian-American youth.
The Journal of Research and Development in Education 27(3), 188–94.
Knight, S., Padron, Y., and Waxman, H. 1985. The cognitive reading strategies of
ESL students. TESOL Quarterly 19(4), 789–91.
Lado, R. 1957. Linguistics across cultures. Ann Arbor, MI: University of Michigan
Press.
Lanauze, M., and Snow, C. 1989. The relation between first- and second-language
writing skills: Evidence from Puerto Rican elementary school children in bilin-
gual programs. Linguistics and Education 1(4), 323–39.
Langer, J. A., Barolome, L., and Vasquez, O. 1990. Meaning construction in school
literacy tasks: A study of bilingual students. American Educational Research
Journal 27(3), 427–71.
Lindholm, K., and Aclan, Z. 1991. Bilingual proficiency as a bridge to academic
achievement: Results from bilingual/immersion programs. Journal of Education
173(2), 99–113.
MacSwan, J., and Rolstad, K. 2003. Linguistic diversity, schooling, and social
class: Rethinking our conception of language proficiency in language minority
education. In C. B. Paulston and G. R. Tucker (eds.), Sociolinguistics: The essential
readings (pp. 329–40). Malden, MA: Blackwell.
P1: JZZ
Maguire, M. H., and Graves, B. 2001. Speaking personalities in primary school

children’s writing. TESOL Quarterly 35, 561–93.
Miramontes, O. 1987. Oral reading miscues of Hispanic students: Implications for
assessment of learning disabilities. Journal of Learning Disabilities 20(10), 627–
32.
1990. A comparative study of English oral reading skills in differently schooled
groups of Hispanic students. Journal of Reading Behavior 22(4), 373–94.
Nagy, W., Garcia, G., Durgunoglu, A., and Hancin-Bhatt, B. 1993. Spanish-English
bilingual students’ use of cognates in English reading. Journal of Reading Behav-
ior 25(3), 241–59.
Nagy, W. E., McClure, E. F., and Mir, M. 1997. Linguistic transfer and the use of
context by Spanish-English bilinguals. Applied Psycholinguistics 18(4), 431–52.
Nguyen, A., and Shin, F. 2001. Development of the first language is not a barrier
to second language acquisition: Evidence from Vietnamese immigrants to the
United States. International Journal of Bilingual Education and Bilingualism 4(3),
159–64.
Nielsen, F., and Lerner, S. 1986. Language skills and school achievement of bilingual
Hispanics. Social Science Research 15, 209–40.
Padron, Y., Knight, S., and Waxman, H. 1986) Analyzing bilingual and monolingual
students’ perceptions of their reading strategies. The Reading Teacher 39(5),
430–3.
Padron, Y., and Waxman, H. 1988. The effect of ESL students’ perceptions of their
cognitive strategies on reading achievement. TESOL Quarterly 22(1), 146–50.
Peregoy, S. 1989. Relationship between second language oral proficiency and read-
ing comprehension of bilingual fifth grade students. NABE Journal 13(3), 217–
34.
Peregoy, S., and Boyle, O. 1991. Second language oral proficiency characteristics
of low, intermediate and high second language readers. Hispanic Journal of
Behavioral Sciences 13(1), 35–47.
Perez, E. 1981. Oral language competence improves reading skills of Mexican Amer-
ican third graders. The Reading Teacher, 35(1), 24–7.
Porter, R. 1990. Forked tongue: The politics of bilingual education. New York: Basic
Books.
Reese, L., Garnier, H., Gallimore, R., and Goldenberg, C. 2000. Longitudinal anal-
ysis of the antecedents of emergent Spanish literacy and middle-school
English reading achievement of Spanish-speaking students. American Educa-
Roberts, T., and Corbett, C. 1997. Efficacy of explicit English instruction in phonemic
awareness and the alphabetic principle for English learners and English proficient
kindergarten children in relationship to oral language proficiency, primary language
and verbal memory. (ERIC Document Reproduction Service No. ED 417 403.)
Rossell, C. H., and Baker, K. 1996. The educational effectiveness of bilingual edu-
cation. Research in the Teaching of English 30(1), 7–74.
Royer, J., and Carlo, M. 1991. Transfer of comprehension skills from native to second
language. Journal of Reading 34(6), 450–5.
Samway, K. D. 1993. This is hard, isn’t it?: Children evaluating writing. TESOL
Quarterly 27(2), 233–57.
P1: JZZ
Literacy 87
Saville-Troike, M. 1984. What really matters in second language learning for aca-
demic achievement? TESOL Quarterly 18(2), 199–219.
Snow, C. E., Burns, M. S., and Griffin, P. 1998. Preventing reading difficulties in young
children. Washington, DC: National Academy of Sciences.
Sulzby, E., and Teale, W. 1991. Emergent literacy. In R. Barr, M. Kamil, P. Mosenthal,
and P. Pearson (eds.), Handbook of reading research (Vol. 11, pp. 727–58). New
York: Longman.
Terrasi, S. 2000. Phonemic awareness skills in kindergarten students from English and
non-English speaking homes (ERIC Document Reproduction Service No. ED 441
220).
Tompkins, G. E., Abramson, S., and Pritchard, R. H. 1999. A multilingual per-
spective on spelling development in third and fourth grades. Multicultural
Education 6(3), 12–18.
Torgesen, J. K., Alexander, A. W., Wagner, R. K., Rashotte, C. A., Voeller, K. S., and
Conway, T. 2001. Intensive remedial instruction for children with severe read-
ing disabilities: Immediate and long-term outcomes from two instructional
approaches. Journal of Learning Disabilities 34, 35–58.
Toukomaa, P., and Skutnabb-Kangas, T. 1977. The intensive teaching of the mother
tongue to migrant children of pre-school age and children in the lower level of com-
prehensive school. Helsinki: The Finnish National Commission for UNESCO.
Urzua, C. 1987. You stopped too soon: Second language children composing and
revising. TESOL Quarterly 21(2), 279–303.
Wagner, R. K., and Torgesen, J. K. 1987. The nature of phonological processing and
its causal role in the acquisition of reading skills. Psychological Bulletin 101,
192–212.
Zutell, J., and Allen, V. 1988. The English spelling strategies of Spanish-speaking
bilingual children. TESOL Quarterly 22(2), 333–40.
P1: JZZ
0521859751c03
88
CB989B/Genesee
table a.3.1. Summary of Studies on L1 Oral Proficiency and L2 Literacy
Sample Outcome Measures

Authors Characteristics Grades Research Design (partial listing) Results
Buriel and 1st, 2nd, 3rd 10, 12 Within gp and btwn – Survey/questionnaire: 1st and 2nd generation:
Cardoza (1998) generation gp comparisons, educational aspirations; Span – Greater L1 oral
0 521 85975 1
Hispanic, various convenience language background (4 pt proficiency and literacy

programs, samples scale for oral and written Span skills than 3rd generation
southwestern – n = approx. 11,300 proficiency, home language – No relationship btwn
USA – Stats, ANOVA; and mother tongue) language background
correlations; – SES variables and reading scores
multiple regression – Standardized reading test 3rd generation:
analysis scores – Those with greater L1
oral proficiency had
lower reading scores
Fernandez and – Eng monolingual High school: Within and btwn gp – Reading and vocabulary test – Proficiency in Eng and
Nielsen (1986) and bilingual (grades comparisons, scores other language positively
Hispanics unspecified) convenience – Self-assessed Eng proficiency related to achievement
– Eng monolingual samples in reading and writing
November 12, 2005
and bilingual – n = 16,046 – Self-assessed Span or other

whites – Stats; regression language proficiency in
– In various analysis reading and writing
programs – Use of other language
15:22
P1: JZZ
0521859751c03
Kennedy and Park Hispanic- and 8 Within gp and btwn – Survey/questionnaire: home – Speaking language other
(1994) Asian- gp comparisons, language background; social than Eng at home
Americans, convenience psychological variables; irrelevant to grades and
nationwide samples student effort standardized reading
sample, various – Hispanic-Americans – Self-reported Eng grades scores for Hispanics
programs (n = 1,952) – Standardized reading test – Negative relationship
CB989B/Genesee
– Asian-Americans scores btwn speaking language

(n = 1,131) other than Eng at home
– Stats: descriptive and standardized test
multiple regression scores in reading for
analysis, Asians
correlations
0 521 85975 1
Lanauze and Span L1 Hispanics, 4, 5 Btwn gp comparisons – Span and Eng teacher-assessed – Children good in Span
Snow (1989) in bilingual of lang proficiency language proficiency level but poor in Eng gp and
programs, New level, convenience (oral, aural and reading skills children good in both
Haven, CT samples combined, but based languages used more
– n = 38 primarily on oral skills; 2 complex and
– Stats, correlations point [good or poor] scale) sophisticated language
– Picture description writing than children poor in
task scored for complexity, both languages
sophistication, and semantic – Children good in both
content languages had
independent writing
skills
– Children good in Span
but poor in Eng
November 12, 2005
transferred skills from

Span to Eng
– Children poor in both
15:22
languages were not
89
transferring skills
(continued)
P1: JZZ

0521859751c03
90
Authors Characteristics Grades Research Design (partial listing) Results
Langer, Barolome, Hispanics from 5 Within gp – Student interviews and school – Students relied on
and Vasquez bilingual homes, comparisons, records: L1 and L2 proficiencies knowledge of Span to
(1990) bilingual convenience – Classroom observation support
program in samples – Passage reading sessions: 2 understanding of Eng
CB989B/Genesee
northern CA – Detailed different genre/text type passages text, increasingly so

ethnographic study (story and report) comparisons; with more difficult
– Case studies during reading “envisionment” texts
– n = 12 questions; post-reading “probing” – Competence in Span
– Stats: descriptive, questions; oral and written enriched reading in
qualitative passage recall L1 and L2
0 521 85975 1
measures – Oral L2 proficiency ratings

Nielsen and Bilingual 12 Within gp – Survey: Eng reading and writing – Language proficiency
Lerner (1986) Hispanics, comparisons, proficiency; Span reading and and reading ability
nationwide, convenience sample writing proficiency; SES, LOR not highly related,
various – n = approx. 1,000 – Reading and vocabulary tests other factors more
programs – Stats, factor analysis scores significant
– No negative effect of
bilingualism on
school achievement
Reese et al. (2000) Span L1 ELLs, Los K-7 Within gp – In-depth home interviews: family – Family literacy
Angeles, CA area comparisons literacy practices (parents use of practices predicted
– Longitudinal Eng or Span literacy at work, emergent Span
– Random sample reading aloud to child) literacy and Eng
November 12, 2005
– n = 66 proficiency, which in
turn predicted Grade
7 reading
15:22
achievement
P1: JZZ
0521859751c03
– Stats: path analysis, – Span early literacy assessment

correlational as (e.g., identify letters and
analysis corresponding sounds, oral
comprehension on story read
aloud, knowledge of print
conventions)
CB989B/Genesee
– Standardized reading tests in

language of instruction (EABE,
CTBS)
– School records
– Standardized tests of Eng reading
performance in Grade 7
0 521 85975 1
– Eng language proficiency:

Bilingual Syntax Measure or Idea
Proficiency Test
– Span proficiency assumed
Saville-Troike ELLs from diverse 2–6 Within gp – Informal parent and teacher – 3 out of 5 top
(1984) L1 backgrounds, comparisons, interviews: home language; achievers used native
L1 literate, from convenience sample personality factors language to figure out
well-educated – n = 19 – Interviews with students in Eng: Eng reading
families, – Descriptive stats, language use; grammatical and
mainstream Eng retrospective content info
program with analysis – ESL classroom observations:
ESL and L1 language use; verbal interaction
instruction – Northwest Syntax Screening Test
(30 min/day) (Eng)
November 12, 2005
– Functional Language Survey (Eng)

– Bilingual Syntax Measure (Eng)
– Reading subtest scores of the
15:22
CTBS
91
P1: JZZ
table a.3.2. Summary of Studies on L2 Oral Proficiency and L2 Literacy
92
0521859751c03a

Authors Characteristics Grade Research Design (partial listing) Results
Goldstein, Hispanic ELLs from 2 7, 8, 9 Within gp comparisons, – Reading comprehension subtest – Significant positive correlation
Harris, schools in 2 districts convenience sample of Peabody Individual btwn adapted story structure
and Klein of southern – n = 31 Achievement Test analysis and reading
(1993) California, in – Stats, correlational analysis – Oral production subscale of comprehension
programs for Language Assessment Scale – Greater relationship btwn
CB989B/Genesee
learning (standardized story retell task), storytelling ability and reading

handicapped, 2 scoring methods: standard comprehension scores than
received bilingual scoring procedure (surface – surface structure analysis and
education in earlier sentence structure and reading comprehension scores
grades vocabulary use in development
of coherent storyline); story
0 521 85975 1
structure analysis (deeper –

types of story structures)
Lindholm Hispanic and English 1, 2, 3, 4 Within and btwn gp – bilingual proficiency, Eng and Grade 3:
and Aclan L1, northern comparisons, convenience Sp scores on Student Oral – High gp significantly outscored
(1991) California, two-way samples: Language Observation Matrix Medium and Low gp on
Span/Eng – Span L1 (n = 159) – CTBS (reading) scores reading scores
immersion (initial – Eng L1 (n = 90) – Eng reading instruction only
reading instruction divided by grade: started in Grade 3
in Span) – Grade 1 (n = 87) Grade 4:
– Grade 2 (n = 71) – High gp performing at grade
– Grade 3 (n = 75) level average in Eng reading
– Grade 4 (n = 16)
and
bilingual proficiency gp:
– High (L2H, L1H)
November 12, 2005
– Medium (L2M, L1M)

– Low (L2L, L1H/M)
– Stats, ANOVA
15:25
P1: JZZ
0521859751c03a
Peregoy Span L1 ELLs, 5 Multiple-case study, comparison – Placement in Eng reading – General correspondence btwn
(1989) transitional of Eng reading proficiency instruction levels by teacher, L2 oral proficiency and L2
bilingual program level gps, convenience based in part on oral reading comprehension
samples: proficiency test scores – Limited vocabulary and
– High (n = 2) – Eng oral language production syntactic knowledge impeded
– Intermediate (n = 2) measure: storytelling from reading comprehension;
– Low (n = 2) 4-frame picture sequence scored however, assistance provided
CB989B/Genesee
– Stats: Descriptive for fluency, semantic content, facilitated reading

grammatical complexity, and comprehension for low gp
grammatical correctness
– Eng reading comprehension
measures of 4 reading passages:
1st read orally; 2nd and 3rd
0 521 85975 1
read silently, followed by

multiple-choice comprehension
questions; 4th read one line at a
time, required to make
interpretations and predictions
after each line
Peregoy Hispanic ELLs, 38 3 Within and btwn gp – Oral and silent reading of L2 reading
and Boyle urban and semirural comparisons of reading gp appropriate L2 passages – Low group demonstrated
(1991) schools in northern levels, convenience samples: followed by multiple-choice extreme difficulty in decoding
CA, bilingual – High (n = 20) questions, and explanations for and comprehending
education and – Intermediate (n = 18) choice – Intermediate gp able to decode
mainstream Eng – Low (n = 19) – L2 oral proficiency assessed and comprehend with difficulty
programs (assigned according to through individual – High gp able to decode and
performance on auditory administration of simulated comprehend with relative ease
vocabulary and word reading science lesson: transcribed and (on par with native Eng peers)
November 12, 2005
subtest of Stanford Diagnostic coded for surface structure

Reading Test) (grammatical complexity and
well-formedness) and deep
15:25
structure (informativeness and
93
comprehension) aspects of L2
oral proficiency (continued)
P1: JZZ
94
0521859751c03a

– Total n = 57 (38 in bilingual L2 oral proficiency
education program, 19 in – Significant difference on all
mainstream Eng program; 25 measures btwn low and high
began reading instruction in reading gps
Eng, 32 began in Span) – Significant differences on
– Stats: not defined well-formedness and
CB989B/Genesee
informativeness btwn
intermediate and high gp
Perez (1981) Hispanic, majority 3 Within and btwn gp with – Prescriptive Reading Inventory – TR (not defined) gp showed
Span L1, Texas unspecified assignment to pre- and post-test significant improvement on
public school, – TR: Instructional Intervention – Instructional Intervention: post-test Inventory compared
program (n = 75) teacher led, oral language to controls
0 521 85975 1
unspecified – CO: regular instruction activities related to concepts in

(n = 75) readers
– Pre- and post-tests
– Unspecified stats
Reese et al. Span L1 ELLs, Los K-7 Within gp comparisons – In-depth home interviews: Grade 7:
(2000) Angeles, CA, area – Longitudinal family literacy practices – Family literacy practices
– Random sample (parents use of Eng or Span predicted emergent Span
– n = 66 literacy at work, reading aloud literacy and Eng proficiency,
– Path analysis, correlations to child) which in turn predicted reading
– Span early literacy assessment achievement
(e.g., identify letters and – Greater oral Eng proficiency
corresponding sounds, oral highly predictive of reading
comprehension on story read performance
aloud, knowledge of print
conventions)
November 12, 2005
– Standardized reading tests in

language of instruction (EABE,
CTBS)
15:25
P1: JZZ
0521859751c03a
– School records
– Standardized tests of Eng
reading performance in Grade 7
Bilingual Syntax Measure or
IDEA Proficiency test
CB989B/Genesee

Royer and Span L1 ELLs 5–6 Within gp comparisons, – L1 and L2 listening – Span reading comprehension
Carlo convenience sample comprehension scores at Grade 5 best predictor of
(1991) – n = 49 – L1 and L2 reading Eng reading comprehension
– Longitudinal comprehension scores at Grade 6
– Stats: correlational and – Eng listening skills second best
regression analyses predictor of Eng reading
0 521 85975 1
Saville- ELLs from diverse L1 2–6 Within gp comparisons, – Informal parent and teacher – Language test scores did not
Troike backgrounds, L1 convenience sample interviews: home language; predict achievement on reading
(1984) literate, from well- – n = 19 personality factors subtest
educated families, – Descriptive stats, retrospective – Interviews with students in – Number of different vocabulary
mainstream Eng analysis Eng: language use; grammatical items used in oral Eng
program with ESL and content info production (interview data)
and L1 instruction – ESL classroom observations: significantly correlated with
(30 min/day) language use; verbal interaction reading achievement, verbosity
– Northwest Syntax Screening did not
Test (Eng)
– Functional Language Survey
(Eng)
– Bilingual Syntax Measure (Eng)
– Reading subtest scores of the
November 12, 2005
CTBS
15:25
95
P1: JZZ
96
0521859751c03a
table a.3.3. Summary of Studies on Component Skills of L2 Literacy Development

CB989B/Genesee

Carlisle et al. (1999) Span L1 ELLs, 1, 2, 3 Within gp comparisons, – Span and Eng receptive – Significant portion of
Chicago, convenience sample vocabulary tests variance in reading
maintenance – Fall and Spring testings (PPVT-R/TVIP) comprehension explained
bilingual program – n = 57 – Test of Auditory Analysis by extensiveness of
0 521 85975 1
– Stats, regression analysis Skills (TAAS) in Eng vocabulary in L1 and L2

– Listening comprehension and phonological
and letter-word awareness
identification tests in Eng – Phonological awareness
– Eng phonological significantly correlated
awareness with Eng vocabulary
– Eng and Span vocabulary
definition task (formal
and informal definitions)
– Eng reading
comprehension (subtest of
CAT)
Cronnell (1985) Span L1 ELLs, some 3, 6 Within and btwn gp – Eng spelling errors in – Significant portion of errors
bilingual classes, comparisons, convenience writing samples classified can be attributed to Span,
Los Angeles, CA sample by type: Span spelling, interlanguage or Chicano
November 12, 2005
– Grade 3 (n = 78) pronunciation, grammar, Eng

– Grade 6 (n = 92) vocabulary
– Stats: descriptive
15:25
P1: JZZ
0521859751c03a
Durgunoglu, Nagy, Span L1 ELLs, 1 Within gp comparisons, – Letter naming ability test – Span phonological
and Hancin-Bhatt transitional convenience sample – Span phonological awareness a significant
(1993) bilingual program – n = 27 awareness test predictor of both Span and
– Stats: correlational – Span and Eng oral Eng word recognition
analysis, multiple proficiency tests (pre-LAS) – Span and Eng oral
regression analysis – Span and Eng word proficiency did not
recognition tests correlate with word
CB989B/Genesee
– Transfer tests: Span to Eng recognition or

word recognition phonological awareness
Fashola et al. (1996) Span L1 ELLs and 2, 3, 5, 6 Within and btwn gp – 40 common Eng words – Significant difference btwn
Eng L1, southern comparisons, convenience selected for Eng/Span Span and Eng, and
CA, program samples contrastive spellings younger and older
unspecified – L1 Span (n = 38) – Eng spelling test students on predicted
0 521 85975 1
– L1 Eng (n = 34) (Span to Eng contrastive

– Stats analysis) errors
Hancin-Bhatt and Span L1 ELLs, 3 4, 6, 8 Within and btwn gp – Background questionnaire – Developmental trend in
Nagy (1994) urban schools in comparisons, convenience – Language use recognition of cognates
predominantly samples: questionnaire compared to non-cognates
Spanish-speaking – Grade 4 (n = 96) – Eng to Span cognate and – Limited knowledge of Span
areas of large – Grade 6 (n = 41) non-cognate translation – Eng systematic
Midwest city, 10 – Grade 8 (n = 59) task relationships btwn suffixes
bilingual – Stats: ANOVA, multiple – Span yes/no vocabulary – Knowledge of Span
classrooms regression analysis test: recognition cognates accounted for
– Eng-Span systematic suffix significant amount of
relationship matching task variance in translation
task; relationship btwn
cognate translation ability
and language
November 12, 2005
background/use factors
(continued)
15:25
97
P1: JZZ
98
0521859751c03a

Hsia (1992) Eng L1 Kindergarten K, 1 Within and btwn gp – Reading readiness test – No significant main effects
and Mandarin L1 comparisons, convenience – Children’s invented – Native-like constraints
K to Grade 1, samples spellings acquired over time
CB989B/Genesee
greater Boston – L1 Eng K (n = 15) – Mandarin phoneme

area, from middle-, – L1 Mandarin K (n = 15) segmentation task
upper-middle class – L1 Mandarin Grade 1 – Eng sentence
homes (n = 15) segmentation task
– Mandarin L1s – 2 testing sessions, 6
attended Chinese months apart
0 521 85975 1
language weekend – Stats, ANOVA

school and
attended/had
attended American
preschool and K
Jimenez, Garcia, and Hispanic ELLs 6, 7 Within and btwn gp – Prior knowledge – Cognate searching strategy
Pearson (1996) (orally bilingual comparisons, convenience assessment background and translating used by all
and biliterate) and samples questionnaire 8 Hispanic successful
Eng L1, 3 schools – Hispanic successful – Teacher and standardized readers
in 2 school readers in Eng (n = 8) test categorization into
districts, some – Hispanic marginally successful and
bilingual schooling successful readers in Eng unsuccessful readers
(n = 3) – Prompted/
– Monolingual Eng unprompted think aloud
successful readers in Eng strategy assessment
November 12, 2005
(n = 3) – Text retellings
– Descriptive stats, – Student interviews
qualitative analysis
15:25
P1: JZZ
0521859751c03a
Nagy et al. (1993) Span L1 ELLs, 2 4, 5, 6 Within gp comparisons, – Multiple-choice test of – Significant difference in
urban elementary convenience sample target cognates cognate over non-cognate
schools, bilingual – n = 74 – Span and Eng yes/no recognition
education and – Stats, MANOVA, multiple vocabulary tests of – Transfer of Span lexical
Eng-only regression analysis target/non-target knowledge to Eng,
programs cognates and dependent on
non-cognates metalinguistic awareness
CB989B/Genesee
– Eng reading of recognizing words as

comprehension cognates; could be
enhanced with
morphological training
Roberts and Corbett ELLs and Eng L1s, K Within and btwn gp – Classroom observation, – ELLs in TR gp scored
(1997) suburban northern comparisons with interviews, family literacy significantly higher on
0 521 85975 1
CA, program convenience assignment interviews some measures of

unspecified – TR: Specific phonological – Eng phonological phonological awareness
instruction in Eng (n = 16 awareness tasks than ELLs in CO gp
L1 Hmong ELLs; 13 L1 – For ELLs – Pre-LAS Eng – ELLs not significantly
Eng) proficiency test different than L1 Eng in TR
– CO: Regular instruction (n or CO gps
= 17 L1 Hmong, 1 L1 Laos – Significant improvement in
ELLs, 11 L1 Eng) Hmong rhyming,
– 2 additional comparison segmenting and blending
classes (n = similar to CO) for Hmong L1s in TR gp
– Pre- and post-testing of all
groups
– Stats, multivariate analysis
Terrasi (2000) Primarily Hispanic K Within and btwn gp – 6 Eng phonological – Significant gains for both
and Eng L1 comparisons, convenience awareness subtests gps
November 12, 2005
students, urban samples – Larger gains for ELLs

schools south of – Primarily Hispanic ELLs
Boston, program (n = 40)
15:25
unspecified – L1 Eng (n = 227)
99
(continued)
P1: JZZ

0521859751c03a
100
– Pre- and post-tested
– Intervention: specific
phonological instruction
in Eng
– Stats: descriptive
CB989B/Genesee
Tompkins, ELLs with diverse 3, 4 Within and btwn gp – Eng spelling errors in – Similar spelling patterns
Abramson, and L1s and Eng L1s, 2 comparisons (language journal writings regardless of language gp
Pritchard (1999) schools, 1 low background, grade and – Significant differences
income and 1 school), random selection: btwn schools
affluent, programs Low-income school – Qualitative analysis
unspecified, – n = 5 (Grade 3), n = 5 showed errors to be largely
0 521 85975 1
central CA (Grade 4), from each lg gp interlanguage

(Sp, Hmong, Lao, Khmer) developmental
(subtotal n = 40)
– n = 5 (Grade 3), n = 5
(Grade 4) L1 Eng
(subtotal n = 10)
Affluent school
– Grades 3 and 4, L1 Eng
(n = 10)
– Stats, ANOVA, qualitative
analyses
Zutell and Allen Span L1 ELLs, large 2, 3, 4 Within gp comparisons, – 5 word categories selected – No differences when
(1988) urban mideastern convenience sample for Span-Eng contrasting grouped by grade
schools, bilingual – n = 108 sound-letter name – When grouped according
programs – Stats: descriptive relationships to test success – less
November 12, 2005
– Eng spelling test successful students

produced more predicted
Span-influenced spellings
15:25
P1: JZZ
0521859751c03a
table a.3.4. Summary of Studies on L1 Literacy and L2 Literacy

Bean, Levine, and Gifted and remedial 11 Btwn gp comparisons, – Graphemic identification – Beginning ESL students paid
Graham (1982) Eng L1s; convenience samples task significantly more attention
intermediate and – L1 Eng gifted to graphemic level of reading
beginner ELLs in ESL (n = 16)
CB989B/Genesee
program, Los – L1 Eng Remedial

Angeles, CA (n = 14)
– Intermediate ESL
(n = 18)
– Beginner ESL
(n = 12)
0 521 85975 1
– Stats: ANOVA
Bermudez and Span L1 ELLs, 2 4 Btwn gp analyses, – Essay samples, written in – No significant difference
Prater (1994) inner-city schools in convenience samples response to standard btwn gps, suggesting that
Southwest USA, ESL – In ESL (n = 18) prompt: persuasive mainstreamed ELLs do not
and mainstream – Already mainstreamed into writing coded for have a higher level of
programs Eng (n = 19) persuasive discourse persuasive discourse needed
– Stats, ANOVA markers to develop as writers
Buriel and Cardoza 1st, 2nd, 3rd generation 10, 12 Within gp and btwn gp – Survey questionnaire: – 3rd-generation students with
(1998) Hispanic, Southwest comparisons, convenience educational aspirations; greater Span literacy skills
USA, various samples Span-language scored higher on reading test
programs – n = approx. 11,300 background (4-pt scale for
– Stats: ANOVA; correlations; oral and written Span
multiple regression analysis proficiency, home
language, and mother
tongue)
November 12, 2005
– SES variables
– Standardized reading test
scores
15:25
101
(continued)
P1: JZZ

0521859751c03a
102
Calero-Breckheimer Span L1 ELLs, major 3, 4 Within gp comparisons, – Line by line reading of – Students used same number
and Goetz (1993) Texan city, bilingual convenience sample Eng and Span texts on of strategies regardless of
education program – n = 26 computer (reading time language, strategy types
– Stats, ANOVA and and lookbacks recorded) highly correlated
correlational analysis – Multiple-choice – More strategies in Eng
comprehension questions positively correlated with
CB989B/Genesee
– After reading free higher scores on

reporting of strategy use comprehension questions
– Strategy use checklist – More strategies in Span,
positively correlated with
more gist recall
Collier (1987) ELLs from 75 different 4, 6, 8, 11 Cross-sectional data from – Eng proficiency upon – Minimum of 2 years of
0 521 85975 1
language 1977–86, age of arrival, arrival schooling in L1 for most

backgrounds, large length of residence, and – Literacy skills upon arrival rapid progress in academic
public school system grade-level comparisons – Number of years of development of L2
on East Coast, ESL – n = 1,548 schooling in Eng Age 8–11:
program – Stats: descriptive – SRA test scores in reading – Achieved grade-level norms
most rapidly
Age 12–15:
– Experienced greatest
difficulty with academic
aspects of L2; probably due
to more complex subject
matter
Field (1996) Span L1 ELLs, 4 Within gp, convenience – Written answers to – Students had difficulty
transitional bilingual sample reading comprehension inferencing, and correctly
classroom, Santa – n = 10 questions interpreting pragmatics of
November 12, 2005
Barbara, CA – Qualitative analysis – Video-taped/audio-taped comprehension questions

gp discussion in Eng and
Span
15:25
P1: JZZ
0521859751c03a
Galindo (1993) Span L1 ELLs, urban 1 Within gp, convenience – Observation, audio – Students interpreted literacy
setting in Southwest sample recordings of classroom events in terms of their own
USA, previously –n=4 literacy events interests and in a manner to
attended bilingual K – Qualitative analysis – Dialogue journals btwn meet teacher’s requirements
writing partners
Jimenez (2000) – Span L1 ELLs, 4, 5, 6 –Classroom and focal – Classroom observations – Emergent findings showed
transitioned to students, convenience – Teacher interviews increased awareness of
CB989B/Genesee
mainstream Eng samples from two Grade 4 – Think-alouds during literacy and basic cognitive
– Eng L1 Hispanic, and one Grade 5 regular reading and interviews operations related to test
mainstream Eng bilingual classes and one with focal students processing
– midwestern city Grade 4–6 bilingual special – Support for linguistically
education class: sensitive, culturally relevant,
– Span L1 ELLs (n = 84) and cognitively challenging
0 521 85975 1
– Eng L1 Hispanic (n = 1) instruction, which helps

– Focal students from each students view dual-language
class (n = 4 – 6) background as a strength
– Teachers (n = 4)
Instructional intervention:
– Bilingual strategic reading
instruction
– Qualitative analysis
Jimenez, Garcia, and Hispanic ELLs (orally 6, 7 Within and btwn gp – Prior knowledge – Cognate searching strategy
Pearson (1996) bilingual and comparisons, convenience assessment background and translating used by all 8
biliterate) and Eng samples questionnaire Hispanic successful readers
L1s, at 3 schools in 2 – Hispanic successful readers – Teacher and standardized
school districts, some in Eng (n = 8) test categorization into
bilingual schooling – Hispanic marginally successful and
successful readers in Eng unsuccessful readers
November 12, 2005
(n = 3) – Prompted/unprompted
– Monolingual Eng successful think-aloud strategy
readers in Eng (n = 3) assessment
15:25
– Descriptive stats, qualitative – Text retellings
103
analysis – Student interviews (continued)
P1: JZZ

0521859751c03a
104
Lanauze and Snow Span L1 Hispanics, had 4, 5 Btwn gp comparison of – Span and Eng – Children good in Span but
(1989) been in bilingual language proficiency level, teacher-assessed language poor in Eng gp and children
program for 1–4 convenience samples proficiency: oral, aural, good in both languages used
years, New Haven, – Good in Span and Eng and reading skills more complex and
CT (n = 17) combined (based sophisticated language than
– Poor in Eng but good in primarily on oral skills; children poor in both
CB989B/Genesee
Span (n = 12) 2-point scale, good or languages

– Poor in Span and Eng poor) – Children good in both
(n = 9) – Picture description languages had independent
– Stats, ANOVA, correlations writing task: complexity, writing skills
sophistication, and – Children good in Span but
semantic content poor in Eng transferred skills
0 521 85975 1
from Span to Eng

– Children poor in both
languages were not
transferring skills
Langer, Barolome, Hispanics from 5 Within gp comparisons, – Student interviews and – Significant main effects for
and Vasquez (1990) bilingual homes, convenience samples school records to assess genre (better understanding
northern California, – Detailed ethnographic L1 and L2 proficiencies of stories over reports) and
bilingual program study – Classroom observation language (Span over Eng);
– Case studies – Passage reading sessions: and type of question
– n = 12 2 different genre/text – Better readers provided more
– Stats, ANOVA, descriptive type passages (story and abstract and
stats, qualitative measures report) comparisons; decontextualized responses;
during reading poorer readers examples and
“envisionment” explanations
questions; post-reading – Those students with good
November 12, 2005
“probing” questions; oral meaning-making strategies

and written passage recall used these strategies in both
– Oral L2 proficiency ratings languages
15:25
P1: JZZ
0521859751c03a
– Good strategies rather than

Eng proficiency
differentiated good and poor
readers
– Competence in Span enriched
reading in L1 and L2
Miramontes (1987) Hispanic, 4 schools in 4, 5, 6 Within and btwn gp – Miscue reading inventory – Both gps of Span readers
CB989B/Genesee
large urban school comparisons, convenience – Graded reading selections adhered significantly more
district in CA, sample – Miscue analysis and closely to the text
Span/Eng bilingual – Good readers in Eng retellings – Good Span readers
program (n = 10) consistently used decoding
– Good readers in Span strategies that adhered more
(n = 10) closely to text in both
0 521 85975 1
– Reading disabled in Eng languages

(n = 10) – Learning disabled in Span
– Reading disabled in Span reading gp did not retain
(n = 9) meaning of text in Eng,
– Stats, ANOVA, Scheffé and suggesting general lack of
factor analysis Eng proficiency – not reading
disability
Miramontes (1990) Hispanics, 2 large 4, 5, 6 Within and btwn gp – Oral reading sessions – Good Span readers
urban school districts comparison, convenience – Miscue analysis significantly paid more
in the Southwest, assignment – Retelling attention to form of text in
Span-Eng full – Good Eng readers: Eng at both languages
bilingual program home, initial literacy in Eng – Good Span readers
(n = 10) successfully used Span
– Good Span readers: initial reading strategies and
literacy Span (n = 10) different strategies in Eng
November 12, 2005
– Mixed-dominance: L1 Span reading from good Eng and

at home, Eng at school Mixed Dominance gps, but
(n = 20) equally effective in
15:25
comprehension
105
(continued)
P1: JZZ
0521859751c03a
106

– Stats, ANOVA, Scheffé, – Good Span readers had
correlational and factor significantly lower scores in
analyses retelling, but may be result of
CB989B/Genesee
more limited oral Eng

proficiency
Nagy, McClure, and Span L1 Hispanic ELLs 7, 8 Within and btwn gp – Language background – Bilinguals influenced by L1
Mir (1997) and Eng L1s, urban comparisons, convenience questionnaire syntactic knowledge when
school district, samples – Eng reading proficiency, guessing meaning of
bilingual and – Span L1 ELLs in bilingual TABE unfamiliar words in Eng
mainstream program program (n = 41) – Span reading proficiency reading context
0 521 85975 1
– Span L1 in Eng mainstream (bilingual program only)

program – Multiple choice, meaning
(n = 45) of nonsense words in Eng
– Eng L1 (n = 15) context
– Stats, ANOVA and
correlational analyses
Nguyen and Shin Vietnamese ELLs, 5–8 Within gp comparisons, – Self-report questionnaire – Near zero correlation of SAT
(2001) program unspecified convenience sample (Likert scale, 16 Qs) of L1 scores and self-report
– n = 170 and L2 competence, competence in L1 literacy
– Stats, rank order correlation preference, attitudes – No evidence that competence
– SAT (reading and in L1 holds back Eng L2
language combined) literacy development
scores
Padron, Knight, and Span L1 ELLs, Eng L1s, 3, 5 Within and btwn gp – San Diego quick – Monolinguals used
November 12, 2005
Waxman (1986) program unspecified, comparisons, convenience assessment graded word significantly more strategies
Knight, Padron, and Houston, TX sample list than bilinguals
Waxman (1985)
15:25
P1: JZZ
0521859751c03a
– Span L1 bilinguals – Think aloud while reading – Gps used different strategies
(n = 23) passages from Ekwall – Monolinguals used
– Eng L1 monolinguals reading inventory concentrating, searching for
(n = 15) details, and self-generating
– Stats, t-tests questions significantly more
– Teacher expectations most
often cited by bilinguals
CB989B/Genesee
Padron and Waxman Span L1 ELLs, small 3, 4, 5 Within gp comparison, – Stanford diagnostic – 2 negative strategies
(1988) industrial town near convenience sample reading test negatively associated with
major southwestern – Pre- and post-testing of – 14-item reading strategy reading gains
city, ESL program reading test and strategy questionnaire – Less successful students used
questionnaire less sophisticated and
– n = 82 inappropriate strategies
0 521 85975 1
– Stats, multiple regression

analysis
Reese et al. (2000) Span L1 ELLs, Los K-7 Within gp comparisons – In-depth home interviews: – Family literacy practices
Angeles, CA, area – Longitudinal family literacy practices predicted emergent Span
– Random sample (parents use of Eng or literacy and Eng proficiency,
– n = 66 Span literacy at work, which in turn predicted
– Stats: path analysis, reading aloud to child) Grade 7 reading achievement
correlational analysis – Span early literacy – Best Span readers earliest to
assessment (e.g., identify transition to Eng reading
letters and corresponding instruction
sounds, oral
comprehension on story
read aloud, knowledge of
print conventions)
– Standardized reading tests
November 12, 2005
in language of instruction
(EABE, CTBS)
15:25
(continued)
107
P1: JZZ
0521859751c03a
108

– School records
– Standardized tests of Eng
reading performance in
CB989B/Genesee
Grade 7
Bilingual Syntax Measure
or Idea Proficiency Test
Royer and Carlo Span L1 ELLs, 5–6 Within gp comparison, – L1 and L2 listening – Span reading comprehension
0 521 85975 1
(1991) transitional bilingual convenience samples comprehension scores at Grade 5 best predictor of
program, Holyoke, – n = 49 – L1 and L2 reading Eng reading comprehension
MA – Stats, correlational and comprehension scores at Grade 6
regression analyses – Eng listening skills second
best predictor of Eng reading
Samway (1993) ELLs, large school 2, 3, 4, 6 Within and btwn gp – Classroom observation – Qualitative analysis showed
district in upstate comparisons, convenience – Audio-taped writing students had awareness of
New York, pull-out samples conferences many facets of writing
ESL classes – Grade 2 (n = 4) – Informal interviews evidenced through their
– Grade 3 (n = 1) – Children’s evaluation of evaluations of writing
– Grade 4 (n = 2) peer and own stories
– Grade 6 (n = 2)
– Qualitative analysis
Urzua (1987) Southeast Asian ELLs, 4, 6 Observational study – Audio-taped process; – ELLs develop skills areas of a
transitioned to –n=2 writing sessions, sense of audience, voice and
November 12, 2005
mainstream feedback, etc. power of language similar to

– dialogue journal writing native Eng-speaking children
15:25
P1: JZZ
Literacy
Instructional Issues
Fred Genesee and Caroline Riches
introduction
Literacy instruction is undoubtedly one of the critical focal points in the
education of all children – native English speakers as well as English lan-
guage learners (ELLs). Literacy is both an end in itself and a means to
other ends since, without formal education, most children would not learn
to read and write and, without reading and writing skills, children would
not be able to learn and function effectively in school and beyond. Clearly,
there are challenges in teaching reading and writing to ELLs that exceed
those that educators face when teaching native English speakers. The focus
of this chapter is on research that has examined the instructional, family
and community, and assessment issues related to reading and writing by
ELLs. For purposes of this review, reading and writing include the production
or comprehension of written language and behaviors related to the production and
comprehension of written language – for example, strategies for comprehending
unknown written words or engagement in reading and writing activities. This
encompasses a broad range of outcome measures, as will become evident in
the following review. The studies were categorized into four broad topics:
1. Instructional Approaches
2. Language of Instruction
3. Family and Community
4. Assessment
instructional approaches
Research reviewed in this section has examined a wide variety of different
methods, techniques, and strategies for promoting the reading and writing
skills of ELLs. Each study was classified according to one of three major
approaches to instruction: (1) direct, (2) interactive, and (3) process-based
109
P1: JZZ
110 Fred Genesee and Caroline Riches
(see Hillocks, 1984, for a similar taxonomy). Briefly, direct instruction

emphasizes the explicit and direct instruction of specific reading/writing
skills or strategies. Interactive instruction emphasizes learning that is medi-
ated through interaction with other learners or more competent readers and
writers (e.g., the teacher). The goals of interactive approaches include spe-
cific literacy skills and strategies, and they also include literacy-related
outcomes including, for example, engagement in reading/writing and
interest in literacy. Finally, process-based instruction emphasizes engage-
ment in authentic use of written language for communication or self-
expression. Process-based approaches deemphasize teaching the compo-
nent skills and strategies of reading and writing in favor of learning through
induction.
These approaches are not mutually exclusive. A given classroom inter-
vention can entail features of more than one approach. For example, to
some extent, all instruction involves the teaching of specific skills and a
social context in which teaching and learning take place; and, all instruc-
tion entails some kind of interaction, at the very least with the teacher. In
fact, many of the studies classified as direct instruction involve interactive
methods. For example, Padron (1992) taught specific comprehension strate-
gies using reciprocal training methods, and Klingner and Vaughn (1996)
used cooperative learning and cross-age tutoring methods. We view the
studies as falling along a continuum of approaches from direct instruction
to interactive to process-based approaches. Some studies had a relatively
distinct constellation of features with a central focus that fitted into one of
these three classifications. Such studies are treated in that category. Some
studies, like Padron (1992) and Klingner and Vaughn (1996), were com-
posed of different approaches with equal emphases. These are discussed
in conjunction with each appropriate approach. Thus, Padron (1992) is
included in the Direct and the Interactive Instructional sections because
it examines classrooms in which a combination of direct instruction and
reciprocal teaching were employed. A handful of studies did not fit into
any of these categories (e.g., a study on suggestopedia, a method that
emphasizes relaxation and the use of music to prepare students to learn
a new language). Such studies were excluded on the grounds that single
studies of unique techniques lack sufficient generalizability to be useful.
The majority of studies examined reading as opposed to writing and stu-
dents in elementary school as opposed to middle or high school. In other
words, most of this research focused on various aspects of the reading
development of ELLs in elementary school.
In the sections that follow, we describe each generic approach in greater
detail, and we review evidence for the effectiveness of each. All relevant
studies are summarized in Tables A.4.1, A.4.2, and A.4.3. Details about stu-
dents, specific instructional methods, research designs, outcome measures,
and results are included. Space limitations do not permit discussion of all
P1: JZZ
Literacy 111
studies, and the reader is referred to the tables for details of studies that
are not reviewed in the text.
Direct Instruction
The studies in this category highlight direct instruction of specific skills that
are considered essential for all students learning to read and write. Direct
approaches to instruction are based on the twin assumptions that reading
and writing consist of interrelated but discrete subskills and that these
skills are best taught explicitly. For example, students are taught new
vocabulary items explicitly or given practice discriminating among sounds
and matching sounds to letters. Another characteristic of direct instruc-
tion is its orientation to evaluation. The effectiveness of direct instruction
is assessed directly and with respect to discrete skills, for example, by
testing students’ vocabulary knowledge or their spelling skills explicitly.
Direct instruction is thought to be particularly appropriate and desirable
for minority language students on the grounds that they are at risk for read-
ing and writing development and, thus, they require explicit and focused
instruction in the requisite skills that comprise reading and writing.
A summary of the studies that are reviewed in this section is presented
in Table A.4.1. Despite arguments in favor of direct instruction (e.g., Snow,
Burns, and Griffin, 1998), our search identified few studies that met our
inclusion criteria – only ten. Research on direct instruction has focused
on reading and paid little attention to writing (except see Bermudez and
Prater, 1990; Echevarria, Short, and Powers, 2003). Within the domain of
reading, text-level skills enjoyed the greatest attention (Bermudez and
Prater, 1990; Hernandez, 1991; Kucer, 1992 [also included in Interactive
Instruction section]; McLaughlin et al., 2000; Padron, 1992; Rousseau and
Tam, 1993), while three studies examined vocabulary skills (Avila and
Sadoski, 1996; McLaughlin et al., 2000; Ulanoff and Pucci, 1999). With
the exception of the Kramer, Schell, and Rubison study, which examined
Hispanic ELLs in Grades 1–3, research has focused on learners in Grades 3
to 7/8. Consequently, there is little empirical evidence concerning the
effects of direct instruction on writing at any grade level or on reading
and writing among early primary school (K-2) or high school ELLs. More-
over, none of the studies examined ELLs whose L1 was not Spanish and,
in particular, ELLs whose native language is typologically distinct from
English, such as Chinese.
All three studies that examined vocabulary report significant improve-
ments in performance (Avila and Sadoski, 1996; McLaughlin et al., 2000;
Ulanoff and Pucci, 1999). More specifically, McLaughlin et al. evaluated the
effectiveness of a vocabulary-enriched curriculum that included “direct
instruction in vocabulary to deepen word knowledge of high-frequency,
grade-appropriate words; instruction in strategies such as how to infer
P1: JZZ
meaning from text, using cognates and recognize root words; and activi-
ties outside the classroom to extend and deepen students’ understanding
of word meanings” (McLaughlin et al., 2000; p. 134). The participants were
Grades 4 and 5 ELLs of Hispanic background and English-only students.
Half of each language group was assigned to the treatment group while
half were assigned to the control group; assignment was based on conve-
nience. The control and treatment group students were drawn from the
same schools, but no description is provided of the former’s instructional
experiences. After two years of exposure to the treatment, ELLs performed
significantly better than control students on measures of knowledge of tar-
get vocabulary, polysemy, morphology, and semantic associations; there
was no significant difference between the groups on the PPVT. Moreover,
the gap between the ELLs and English L1 students was attenuated by
40 percent after two years of exposure to the treatment condition. How-
ever, the ELLs continued to score significantly lower than the English L1
students.
In the Avila and Sadoski (1996) study, Grade 5 ELLs of Hispanic back-
ground were taught new English vocabulary using a Spanish keyword
method. Students in the control condition were taught new English words
using a translation method. Students exposed to the keyword method
demonstrated significantly superior word-knowledge skills in English in
comparison to the control group. The advantages of the treatment were
evident immediately following intervention and after a delay of one week.
These findings also attest to the effectiveness of cross-language skills train-
ing, an issue we return to later. Finally, Ulanoff and Pucci (1999) found
that Grade 3 ELLs exposed to a preview-review method of vocabulary
development had greater post-treatment scores (m = 14.87) than did ELLs
who had been exposed to a concurrent translation treatment (m = 7.33)
and control ELL students (m = 10.44). In fact, the concurrent-translation
group scored significantly lower than the control students. There were no
significant differences between groups at pre-test.
We identified only three studies that analyzed the effects of direct
instruction on English text-level reading skills that used statistical pro-
cedures (Bermudez and Prater, 1990; McLaughlin et al., 2000; and Padron,
1992); among these, two reported significant advantages (McLaughlin,
2000; Padron, 1992) and one reported nonsignificant advantages (Bermu-
dez and Prater, 1990) for students who received direct instruction. Three
additional studies in this group provide narrative descriptions of the
effects of direct instruction on reading performance (Hernandez, 1991;
Kucer, 1992; Rousseau and Tam, 1993) and, thus, must be interpreted with
reservation.
Padron (1992) examined the effects of direct instruction on Grades 3 to 5
Hispanic students’ use of specific comprehension strategies: (1) question
generating, (2) summarizing, (3) predicting, and (4) clarifying. There were
P1: JZZ
Literacy 113
two treatment groups, each of which entailed twice-weekly, 30-minute

sessions for one month; one treatment group was given reciprocal train-
ing and the other question-answer–relations training. In the reciprocal-
training group, teachers discussed why some students experience diffi-
culties understanding text. This was followed by sessions during which
the teacher modeled the four targeted comprehension strategies, follow-
ing which students were given opportunities to use the strategies. In the
question-answer group, the targeted strategies were taught by having
students classify comprehension questions according to what strategies
they could use to arrive at an answer. Students’ use of the strategies was
assessed by the Reading Strategy Questionnaire (RSQ), a self-report meas-
ure. Students in Control Group 1 were introduced to the same story as
that read by the students in the treatment groups, and then read it silently,
with none of the special training. Students in Control Group 2 took the
RSQ but had no special intervention; they then remained in the class and
received instruction in a subject other than reading. Control 2 assessed the
effects of prior exposure to the questionnaire on post-test performance.
In response to the RSQ, students in the reciprocal-teaching group indi-
cated that they used more “strong” reading strategies (i.e., “summarizing”
and “self-generated questions”) than the control students, while students
in both treatment groups reported using fewer “weak” strategies (e.g.,
“thinking about something else” and “writing down every word in the
story”) than the control students. There were no differences between the
treatment and control groups on the remaining ten strategies surveyed in
the questionnaire. The design of this study is exemplary in its inclusion of
alternative treatment and control groups. However, the question remains
whether their self-reported increase in the use of certain strong strategies
was actually associated with greater use and whether it resulted in greater
comprehension; this was not examined by Padron. This is a more general
issue of relevance to studies of strategy use; namely, whether increased
use of specific strategies that are the object of direct instruction actually
improves reading comprehension (see also Kucer, 1992).
Rousseau and Tam (1993) examined the decoding and comprehen-
sion skills of eleven- to twelve-year-old Hispanic students with speech/
language impairment following direct instruction using a keyword method
alone, a listening preview method alone, or a combined keyword plus lis-
tening preview method. Student performance was scored numerically, but
there were no statistical analyses of their results and, thus, these findings
must be qualified accordingly. Although the lack of statistical analyses
limits the significance of these results, this small sample study (n = 5) is
noteworthy for its research design. Each student was observed initially dur-
ing a baseline condition, followed, in order, by keyword alone, listening
preview alone, and combined keyword plus listening preview treatments.
This design allowed the researchers to examine the relative effects of these
P1: JZZ
two types of intervention, alone and in combination, holding individual

student differences constant. This study is also noteworthy for its attention
to ELLs with speech and language impairment, a group that is not well rep-
resented in current research. The keyword method alone produced better
performance than listening preview alone but not as high as the combined
treatment condition.
Bermudez and Prater (1990) examined the effects of direct instruction
on both reading comprehension and writing in Grades 3 and 4 ELLs of
Hispanic, low SES backgrounds. Writing performance was assessed using
measures of fluency, elaboration of ideas, and organization of ideas. Stu-
dents’ comprehension was assessed post-treatment by a series of questions
that tapped their literal and inferential understanding of three selections
from their basal reader. Students in the treatment condition first read three
selections from their basal reader; they then engaged in brainstorming and
clustering of ideas about the topic of the stories, with the assistance of the
teacher; and, finally, they wrote a paragraph about the topic. The inter-
vention was brief, lasting only two days. Students in the control condition
also read the stories from the basal reader and wrote a paragraph about
them. They too were engaged in a teacher-led discussion of the stories
prior to writing about them; but, in contrast to the treatment condition,
their discussion was based on questions from the basal reader. Students in
the treatment group received significantly higher scores than the control
group on elaboration of ideas in their writing; but there were no differences
between the groups on written fluency and organization. Moreover, there
was no significant difference between the treatment and control group on
the comprehension measure.
McLaughlin et al. (2000), as reported previously, found that Grade 5
ELLs who had been taught using a vocabulary-enriched curriculum for
two years scored significantly higher on a cloze/comprehension test than
ELLs who had not had this instruction. Moreover, the gap in reading com-
prehension between the ELLs and English L1 students was reduced from
1.06 standard deviations after year one of the project to 0.50 standard devi-
ations at the end of year two, a reduction of approximately 45 percent.
A number of additional studies have examined reading comprehen-
sion but provide only narrative descriptions of student performance
(Hernández, 1991; Kucer, 1992, 1995; Rousseau and Tam, 1993). We have
included these studies in our review despite this limitation because of the
paucity of studies on direct instruction and in order to give a comprehen-
sive overview of extant work with this focus. All three of these studies
report post-treatment improvements in ELLs’ reading performance, but
these findings must be interpreted with caution. Hernández (1991) found
that instruction in reading-comprehension strategies delivered in Spanish
to Hispanic students in the summer prior to Grade 7 resulted in statisti-
cally significant gains in their reading comprehension in Spanish and, most
P1: JZZ
Literacy 115
important also resulted in use of these strategies during English reading.

Because Hernández’s analysis of the transfer data is only descriptive and
the data were based on only seven students, the results must be interpreted
cautiously. These results, along with those from Avila and Sadoski, are nev-
ertheless theoretically and practically significant because they address the
issue of crosslanguage transfer of reading comprehension strategies. More
research is clearly called for to examine this possibility further.
Kucer (1992) used modified cloze reading lessons to teach Mexican-
American ELLs how to use context clues when they encounter words they
do not know how to read. The intervention took place over the course
of the Grade 3 school year. Kucer observed that 93 percent of the stu-
dents’ responses to unknown words in the cloze passages were contex-
tually appropriate. However, interviews with the students indicated that,
despite their good performance on the cloze passages, they often misun-
derstood the teachers’ intent in using them. Kucer cautions that it is not
sufficient to directly teach reading and writing strategies. Students also
need explicit explanations of how and why they are useful, and they must
be motivated to use them.
The effects of direct-skills training have also been examined on auditory
discrimination (Kramer, Schell, and Rubison, 1983) and writing (Bermudez
and Prater, 1990; Echevarria, Short, and Powers, 2003). The Kramer et al.
study is included on the premise that auditory discrimination is a precur-
sor to acquiring sound-letter knowledge and early decoding skills. Kramer
et al. (1983) examined the effectiveness of a four-week auditory discrim-
ination training program in English for Grades 1 to 3 Spanish-speaking
ELLs. The premise of the intervention was that poor discrimination skills
may hamper Spanish-speaking students’ initial decoding skills in English.
The sound pairs in English that were taught during the training sessions
were selected because they are difficult for native speakers of Spanish.
Post-treatment testing included contrasts that were taught as well as some
that were not taught. Control students received none of the discrimination
training. The treatment students discriminated significantly better than
the control students on all sound pairs that were taught and, as well, on
additional pairs that were not taught.
Echevarria et al. (2003) compared the writing skills of Grades 6, 7, and 8
ELLs with diverse L1s following one year of instruction by trained teachers
using the Sheltered Instruction (SIOP) method developed by Echevarria,
Vogt, and Short (2000). The end-of-year writing scores of the SIOP students
were compared to those of a similar group of ELLs. Analysis of covariance
with beginning-of-year scores as the covariate were conducted to com-
pare the students’ results at the end of the school year. There were statisti-
cally significant differences in favor of the SIOP students on total writing
score, language production, organization, and mechanics; there were no
between-group differences on the focus and elaborations scores. As noted
P1: JZZ
previously, Bermudez and Prater (1990) report advantages for ELLs fol-
lowing direct-writing instruction using brainstorming and clustering on
elaboration but not on fluency or organization.
Interactive Instruction
A recurrent issue in literacy education concerns the nature of the broader
social context in which students learn to read and write. A number of the-
oreticians emphasize the importance of interactive learning environments
to promote reading and writing proficiency (e.g., Cummins, 1984; Slavin,
1995; Tharp, 1997). In interactive learning environments, learners engage
in literacy activities with one or more other learners or with more mature
readers and writers (like teachers, parents, or older students). In this way,
students learn from others, initially by observation and subsequently by
internalizing more mature literate behaviors exhibited by others. In con-
trast to learning in direct instruction, learning in interactive instructional
environments is indirect or mediated by such social interaction. Interac-
tive approaches are favored by some on the grounds that teachers and
parents who are competent readers and writers can provide learners with
individualized guided instruction that corresponds to their zone of prox-
imal development, in line with Vygotsky’s theory of development and
learning.
Some researchers have argued that interactive learning environments
are especially relevant to ELLs because of the diverse sociocultural back-
grounds of these students. More specifically, interactive approaches sup-
port individualized teaching and learning in line with the heterogeneous
learning needs and styles of ELLs. Interactive learning environments are
also thought to reinforce participant structures that some ELLs are used
to in their homes but which differ from mainstream American culture.
These participant structures emphasize group versus individualized par-
ticipation, collaborative versus competitive demonstrations of compe-
tence, and learning by observing versus learning by talking. Interactive
learning environments entail multiple participants engaged in collabo-
rative work and, consequently, extended opportunities to learn through
observation. Learning from models is also thought to be advantageous
for students from minority-language backgrounds who have not had
extensive extracurricular experiences with adult models of literacy; the
same could be said of majority-language students from low-literacy back-
grounds. A further argument in favor of interactive approaches comes from
the notion that reading and writing are more than mere cognitive activi-
ties. They are linked to a culture of literacy (Hudelson, 1994). Interactive
strategies recognize and promote the acquisition of this culture in addi-
tion to the specific language/cognitive skills that comprise reading and
writing as cognitive activities. Descriptions of the learning environments
P1: JZZ
Literacy 117
table 4.1. Sample Descriptions of Interactive Learning Environments
e.g. 1: from Calderon et al. (1998, p. 157) “teachers assign students to four–
member, heterogeneous learning teams of students, who work together to help
each other learn . . .” “ . . . the interaction and practice with peers helped
students develop fluency and comfort with English.”
e.g. 2: from Fayden (1997, p. 25): “The teacher was the model for reading. As she
and the children reread the book many times over a period of several days, the
teacher gradually withdrew herself as the children assumed more and more
responsibility for reaching the book.”
e.g. 3: from Klingner & Vaughn (1996, p. 276): “At first, the teacher models use of
these strategies by ‘thinking aloud’ as she reads through a text. The teacher
then leads students in a text–related discussion, assisting them in strategy use
and gradually withdrawing support as it is no longer necessary. As students
become more proficient . . . they take turns being the ‘teacher’ and leading
discussions about text content.”
e.g. 4: from Klingner & Vaughn (2000, p. 70): “. . . it is based on the theory that
cognitive development occurs when concepts first learned through social
interaction become internalized and made one’s own.”
of a number of the studies reviewed in this section are presented

in Table 4.1.
In keeping with the broad range of goals of interactive approaches to
literacy instruction, a wide variety of specific interactive teaching/learning
environments has been implemented and examined in the literature (see
Table A.4.2). The diversity of interactive techniques for literacy instruc-
tion exhibited in this body of research is, in turn, reflected in a diversity
of reading, writing, and literacy-related outcome measures. Table A.4.2
provides a description of the outcomes examined in each study that were
relevant to our review. A number of the studies used norm-referenced
measures of general reading; others used discrete-point tests of specific
reading skills related to vocabulary, letter identification, and the lexical
and propositional content of written text, to give but a few examples. Still
others examined students’ use of reading strategies and reading-related
behaviors, such as helping, engagement in reading, perceptions of control
of reading, and interests and attitudes. The latter are termed “reading-
related” in this review since they assess reading indirectly. Some teachers
and literacy specialists argue that such behaviors, while ancillary to read-
ing per se, are important components of a developmentally appropriate
program of reading instruction. They argue further that these ancillary
skills are particularly important in instruction for ELLs who have no or
limited exposure to literacy outside of schooling and, thus, require instruc-
tion that attends to the broader context of reading; the same could be said
of native English speakers with limited literacy experiences before coming
to school.
P1: JZZ
The majority of studies examined text-level reading comprehen-

sion skills (Calderon, Hertz-Lazarowitz, and Slavin, 1998; Cohen and
Rodriquez, 1980; Doherty et al., 2003; Echevarria, 1996; Goldenberg,
1992/93; Klingner and Vaughn, 1996; Saunders and Goldenberg, 1999; and
Syvanen, 1997), including the use of strategies related to reading com-
prehension (Fayden, 1997; Klingner Vaughn, 1996; Padron, 1992). Three
studies examined vocabulary comprehension skills (Doherty et al., 2003;
Klingner and Vaughn, 2000; Kucer, 1992), and only two examined writing
(Calderón et al., 1998; Goldenberg, 1992/93). A number of studies exam-
ined other reading-related behaviors, including reading fluency and accu-
racy (Blum et al., 1995; Li and Nes, 2001), engagement in reading (Blum
et al., 1995; Klingner and Vaughn, 2000; Martinez-Roldan and Lopez-
Robertson, 2000, also included in Process Approaches), and use of aca-
demic discourse (Echevarria, 1996). With the exception of Echevarria, all
other studies that examined reading-related behaviors provide narrative
descriptions of their results and, thus, must be interpreted with caution.
With the exception of Echevarria (1996) and Klingner and Vaughn (1996),
both of whom examined middle-school students, all other studies exam-
ined students in elementary school. There were no studies of pre-school or
high school students.
A number of general trends emerge from these studies. First, it appears
that interactive instructional strategies can be effective with ELLs, as
argued by its advocates. Virtually every study in this corpus reported that
ELLs in interactive learning environments demonstrated improvements in
reading and writing or behaviors related to reading and writing as a con-
sequence of participation in an interactive learning environment. This was
shown in studies that used between-group designs (Calderón et al., 1998;
Padron, 1992), within-group (pre-post) designs (Fayden, 1997; Klingner
and Vaughn, 1996, 2000), and regression designs (Doherty et al., 2003). It
was also shown in studies that provide only narrative descriptions with
no numeric or statistical data (Blum et al., 1995; Kucer, 1992; Li and Nes,
2001; Martinez-Roldan and Lopez-Robertson, 2000). A number of these
narrative reports also suffer from small sample sizes – five or less in some
studies (Blum et al., 1995, Kucer, 1992; and Li and Nes, 2001).
The exceptions to the overall pattern of improvement following interac-
tive lessons were Syvanen (1997), Goldenberg (1992/93), and Cohen and
Rodriquez (1980). More specifically, Syvanen (1997) found no significant
difference in improvement from pre- to post-test among Grades 4 and 5
ELLs who had participated in a cross-age tutoring treatment on a district
reading achievement test in comparison to a sample of seventy students in
regular classes elsewhere in the district. Without proper control measures,
however, it is difficult to know if the comparison group was comparable in
other respects to the ELLs who were the subjects of this study. Goldenberg
(1992/93) reports no significant differences in literal comprehension
P1: JZZ
Literacy 119
between students who had participated in instructional conversation (IC)

lessons in comparison to students who had had basal reading (BR) lessons,
although he reported that the IC group demonstrated more complex and
sophisticated conceptualizations of the theme of the reading lesson in their
written essays.
As well, Cohen and Rodriquez (1980) report no advantage for Grade 1
Hispanic ELLs who participated in interactive group-oriented classrooms.
The latter study warrants some discussion since it compared the reading
achievement of students exposed to two contrasting modes of instruction –
high intensity (direct) skills instruction (HIL) and group-oriented interac-
tive instruction. This is a powerful design since it serves to evaluate the
impact of different approaches rather than simply show that a specific
approach can result in improvement. Cohen and Rodriquez (1980) found
that ELLs in HIL classrooms demonstrated higher reading comprehension
scores on the CTBS than did students in the interactive classrooms.
Notwithstanding these exceptions, this corpus of studies indicates that
interactive instruction can enhance reading comprehension skills related
to both vocabulary and text-level material, writing skills (although the evi-
dence here is scant), and other reading-related behaviors, as described pre-
viously. Again, however, it is important to keep in mind that the research on
these other behaviors must be interpreted cautiously owing to its narrative
reporting procedures.
The Saunders and Goldenberg (1999) study warrants some discussion
here because it exemplifies a “component building” approach to instruc-
tional development and evaluation that could serve as a model for others.
Specifically, they examined the effects of two instructional components –
IC and literature logs (LL), on the performance of Grades 4 and 5 ELLs
of Hispanic background. Students were randomly assigned to one of four
groups: LL only, IC only, IC+LL, or control. Students’ were tested for their
factual and interpretive comprehension and for their understanding of
the themes of a story about “giving.” In the LL lessons, teachers met the
assigned group of students briefly, gave them prompts, and asked them to
write about personal experiences that were related to the story. Students
then wrote learning logs independently, and the teacher subsequently
engaged them in a discussion about the similarities and differences in their
experiences and those of the characters in the stories. In the IC lessons,
“teachers attempted through discussion to clarify the factual content of
the story and develop students’ understandings of the more sophisticated
concept of giving . . .” (Saunders and Goldenberg, 1999, p. 287). Students in
the control group did not participate in small group discussions with the
teacher. Instead, they worked alone or with a teaching assistant on read-
ing and writing activities related to social studies. Students in the LL +
IC group scored significantly higher than the control group on story com-
prehension, and students in all three treatment groups were more likely
P1: JZZ
to demonstrate an understanding of the story themes than students in the

control group.
There was an interesting and important interaction effect for the “under-
standing of story theme” measure that involved the ELLs and English-
only students. In particular, English-only students benefited from all three
treatments, whereas only ELLs in the combined LL+IC group benefited
significantly (statistically); in other words, there were no significant differ-
ences between ELLs in the IC-only and LL-only groups in comparison to
the control group. With respect to factual and interpretive comprehension,
all students showed enhanced performance in the combined condition in
comparison to all other conditions. Overall, the effects of IC were some-
what stronger than those of LL, arguing that given a choice, teachers are
advised to choose the former. Implementation of the combined approach
is recommended otherwise since it clearly benefits ELLs the most (see also
Rousseau and Tam, 1993).
A second trend to emerge from this group of studies is the effec-
tiveness of interactive approaches with ELLs from a variety of back-
grounds, more specifically, with ELLs from low socioeconomic status
(SES) families (Doherty et al., 2003; Fayden, 1997; Padron, 1992; Golden-
berg, 1992/93; Saunders and Goldenberg, 1991, 1999), ELLs with learn-
ing disabilities (Echevarria, 1996; Klingner and Vaughn, 1996), and ELLs
with emergent literacy skills (Blum et al., 1995). The findings from Blum
et al. must be interpreted with caution owing to the descriptive nature
of their data. There is also evidence that interactive learning environ-
ments benefit ELLs from diverse ethnolinguistic backgrounds: Native
American (Fayden, 1997), Chinese American (Li and Nes, 2001), and His-
panic American students (e.g., Calderón et al., 1998; Doherty et al., 2003;
Echevarria, 1996; Goldenberg, 1992/93; Klingner and Vaughn, 1996, 2000;
Kucer, 1992; Padron, 1992; Saunders and Goldenberg, 1999). Li and Nes’s
results on Chinese Americans are based on narrative descriptions only and,
thus, must be interpreted with caution. Overall, students of Hispanic back-
ground were the focus of attention in most studies. In light of the apparent
language socialization differences among families from different cultural
backgrounds (Ochs and Schieffelin, 1984), more carefully designed, in-
depth research is warranted on the effectiveness of interactive approaches
with learners of other backgrounds.
Third, interactive approaches also appear to be effective with ELLs in
middle school (Echevarria, 1996; Klingner and Vaughn, 1996) as well as in
elementary school (see Table A.4.2 for those studies). However, as noted
previously, studies on ELLs in high school are lacking. This gap is of par-
ticular concern given the critical role that reading and writing play in the
mastery of academic subjects, such as mathematics and science, in the
higher grades.
P1: JZZ
Literacy 121
Finally, interactive approaches appear to be effective in promoting

reading-related behaviors; that is, engagement in reading and writing and
an understanding and appreciation of literacy in its broader sense, as well
as text- and word-level comprehension skills. As noted previously, how-
ever, all of the studies that examined reading-related outcomes provide
only narrative descriptions, arguably owing to the general and complex
nature of these outcomes.
Process Approaches
Process approaches emphasize student engagement in authentic literacy
activities with significant communicative goals. Typically, students are
given extended opportunities to engage in free reading or writing and
in reading and writing activities in which communication is emphasized,
such as dialogue journals, literature logs, or free reading/writing. Engage-
ment in reading and writing activities may be individual or interactive –
dialogue journals or free writing, for example, are usually individual
activities, whereas shared literature can entail group activity. Children’s
literature is a common vehicle for implementing process approaches since
literature exposes learners to authentic written text, is engaging, and allows
learners to relate to written language via their own experiences, if materials
are well chosen. As Roser, Hoffman, and Farest (1990) indicate, literature-
based literacy programs provide a number of advantages to ELLs: they
“(1) offer exposure to a variety of children’s books, (2) contribute to a rich
literary environment, (3) motivate responsive reading, (4) encourage vol-
untary reading, (5) expand the learners’ reading interests, (6) help learners
grow in language, reading, writing, and thinking, and (7) help learners
discover their own connections with literature.”
Process approaches are distinguished by the view that language is holis-
tic – reading, writing, speaking, and listening (as well as their component
subskills) co-occur under authentic conditions and they, therefore, should
be taught and learned together. See Table 4.2 for some sample descriptors of
process-based instruction in studies reviewed in this section. Proponents of
the process approach view the distinctions between the subcomponents
of reading and writing that are emphasized in direct skills-based
approaches and even some interactive approaches as artificial. Moreover,
they argue that focusing instruction on subskills is less likely to succeed
because it focuses students’ attention unduly on the component elements
of literacy while distracting them from the ultimate goal – reading and
writing for authentic communication and self-expression. This is not to say
that process approaches are indifferent to the mastery of spelling, grammar
skills, and so forth rather, they view the acquisition of these subskills as
a natural by-product of engagement in communicatively oriented reading
P1: JZZ
table 4.2. Sample Descriptions of Process Approaches
• e.g. 1: from Gomez et al. (1996, p. 218): “Students in the Free Writing group . . .
selected their own topics, and could write for as long as they wanted. Students’
writing was not subjected to error corrections . . . teachers responded to each
student’s writing through written comments. Students were then invited to
respond . . ., thus creating a written dialogue. In the Structured Writing
group . . . topics were assigned by the teacher, and the students wrote
intensively, in nine minutes of concentrated writing time. Students were
instructed to work alone and quietly . . . Writing samples were subjected to error
correction by the teacher. . . . Students were directed to focus on avoiding those
errors on their next writing sample.”
• e.g. 2: from Kuball Peck (1997, p. 217): “The classroom was a print-rich
environment in which skills were learned in context as part of a whole. For
instance, the teachers modeled reading and writing on a daily basis. Recipes,
songs, stories, and daily news were charted in front of the students.
Child-dictated stories were transcribed by the morning teacher . . . thus, skills
were presented in context. Fragmented instruction, in which skills are taught in
isolation, was not offered.”
and writing. In fact, most evaluations of process approaches included in

our review use direct assessments of reading and writing skills (see, for
example, Roser et al., 1990; Schon, Hopkins, and Davis, 1982; and Schon,
Hopkins, and Vojir, 1984). The question is, how effective are they at pro-
moting acquisition of specific reading and writing skills in the absence of
focused or direct instruction in such skills?
Whole language can be viewed as a special case of the process approach
since it shares these tenets of other process approaches. Indeed, a defining
characteristic of whole language is its emphasis on the integrity of reading,
writing, speaking, and listening (and their respective subskills). Whole-
language philosophy asserts that the acquisition of literacy skills occurs
naturally, like the acquisition of oral and aural language, through involve-
ment in authentic, meaningful uses of written language. There is wide
variation among whole-language programs with respect to the instruction
of the component skills of reading and writing.
As was the case for the other approaches discussed in this chapter, there
are a number of ways in which process approaches are conceptualized,
operationalized, and evaluated (de la Luz Reyes, 1991). Table A.4.3 summa-
rizes the variety of instructional techniques and foci that were investigated
and the outcome measures used to evaluate them in the studies included
in this review. The description of the outcome measures in Table A.4.3 is
not intended to be complete. Rather, it illustrates the wide range of out-
come measures that have been used in this research, in keeping with each
program’s particular conceptualization of the approach. To be more spe-
cific, a number of socioaffective variables (e.g., attitudes toward reading;
P1: JZZ
Literacy 123
self-concept as a reader/writer; engagement in reading) figured in a num-

ber of these studies along with more conventional outcomes measures,
such as spelling, grammar, and standardized test scores. Proponents of
process approaches regard it as the preferred method of instruction for
ELLs on the assumption that they are particularly responsive to the special
language learning needs of ELLs.
Like research on the other approaches, the research in this corpus
focused on word level (Schon et al., 1982) and text-level reading skills or
reading-related behavior (Carger, 1993; Kucer and Silva, 1999; Kucer,1995;
Martinez-Roldan and Lopez-Robertson, 2000; Roser et al., 1990; Schon et al.,
1982; Schon et al., 1984). In contrast, there was also a focus on the devel-
opment of writing (Carger, 1993; de la Luz Reyes, 1991; Gomez, Parker,
and Lara-Alecio, 1996; Kuball and Peck, 1997; Kucer and Silva, 1999). All
of the studies examined students in elementary grades, except Schon et al.
(1984), who examined students in high school (Grades 9–12). ELLs of
Hispanic backgrounds were the only student group to be investigated.
Overall, evidence of the effectiveness of process approaches is mixed.
Six studies report advantages for students exposed to process instruction
(Carger, 1993; Gomez et al., 1996; Kuball, and Peck, 1997; Kucer and Silva,
1999; Martinez-Roldán and Lopez-Robertson, 2000; Roser et al., 1990),
but with the exception of Kucer and Silva, all other studies provide only
narrative descriptions with no statistical analyses of their results; and two
report results only for reading-related behaviors (Carger (1993): expres-
sion of emotion during pretend reading in one child; Martinez-Roldán and
Lopez-Robertson (2000): engagement in reading. Two studies report no
statistically significant advantages on standardized tests of reading (i.e.,
Inter-America Reading Test, MAT) for ELLs who experienced process-
based literacy activities in comparison to control students (Schon et al.,
1982; Schon et al., 1984). Schon et al. (1982) examined Hispanic ELLs from
low SES families who were in Grades 2–4, while Schon et al. (1984) exam-
ined Hispanic ELLs from low SES families in Grades 9–12. Yet others report
less favorable outcomes for students in process-based literacy classrooms
(de la Luz Reyes, 1991; Gomez et al., 1996). Of particular note, Gomez et al.
(1996) report that Hispanic ELL students who received “structure-based
lessons” in the summer prior to entering Grade 6 outperformed students
who received extended opportunities for free writing, an activity that is
often associated with process instruction.
Evidence for the effectiveness of process approaches to literacy instruc-
tion is even more tentative when careful consideration is given to method-
ological factors. While these studies vary with respect to the detail and thor-
oughness of the descriptions they provided of the instructional approaches
under investigation, overall, many suffer from inadequate descriptions of
the actual literacy activities. This is especially true of those that examined
whole-language classrooms, but it is not only these studies that suffer such
P1: JZZ
problems. For example, Schon et al. (1984) note that “Teachers in the exper-
imental group were instructed to provide at least sixty minutes a week of
free reading time and to do everything they could to help their students develop
positive attitudes towards reading.” (emphasis added) (p. 14).
A number of researchers, even those who argue for a process approach,
called for a balanced approach that incorporates some direct instruc-
tion of specific skills, as needed, embedded in process-based activities.
For example, Kucer and Silva (1999), noted earlier, comment that “. . . it
is overly simplistic to assert that students will improve their literacy
abilities by being immersed in a garden of print; that is, students will
improve in their reading and writing due to the maturation process, regard-
less of instruction. . . .” (p. 365). A similar conclusion is drawn by de la
Luz Reyes (1991) following a study of the writing abilities of Grade 6
Hispanic ELLs in classrooms where dialogue journals and literature logs
were used to promote development: “Overall, mere exposure to standard-
ized writing conventions did not improve the students’ use of them”
(p. 291). In response to this situation, Kucer and Silva recommend that
“. . . when it is determined that a child is encountering repeated difficulty
with a particular dimension of written language, focused instructional
events would be developed that explicitly teach over time the matter
in which the child is experiencing difficulty. In these lessons, not only
is the child shown what to do, but also how it is to be accomplished”
(p. 366).
Clearly, considerably more research is necessary to come to firm con-
clusions about process approaches. In particular, more research is called
for that identifies the critical features of this approach as well as the nec-
essary conditions for successful implementation of such approaches. In
the meantime, current evidence suggests that process approaches alone
are not particularly effective at promoting the acquisition of reading- and
writing-specific skills unless provision is made for such a focus.
Methodological Considerations
Methodological concerns that are particularly relevant to specific
approaches have been discussed in the preceding sections. Here we address
methodological issues that emerge from a consideration of all three bod-
ies of research. There is clearly variation in the methodological strength
of individual studies. In their favor, many studies are noteworthy for
the detail and care that was taken to provide descriptions of the actual
implementation of specific instructional strategies. Doherty et al. (2003)
and Echevarria (1996), moreover, assessed the fidelity of implementa-
tion of the IC approach that was the focus of their investigations. While
highly recommended, this is not a common practice in this entire body of
research.
P1: JZZ
Literacy 125
Future research on the effectiveness of alternative instructional

approaches would benefit from a number of methodological improve-
ments. Systematic use of objective measures of learner outcomes and
appropriate statistical analyses of these outcomes is critical to determine
the magnitude and reliability of instructional effects; a number of studies
noted in summary Tables A.4.1, A.4.2, and A.4.3 provided only narrative
descriptions. As well, some studies used small sample sizes, making gen-
eralizability difficult. Small sample sizes can be justified in ethnographic
studies that entail in-depth descriptions of student involvement in reading
and in studies of special populations that are difficult to identify (e.g., ELLs
with learning/language disabilities [see Echevarria, 1996, for an example]).
At the same time, follow-up studies with larger samples are called for to
confirm trends noted in such studies. None of the small-n studies reported
such follow-up research. Long-term as well as short-term assessment of the
impact of instruction is also highly desirable, for obvious reasons. In the
majority of studies reviewed here, students from Hispanic/Latino back-
grounds were examined, leaving open to question the generalizability of
findings to other groups. Clearly, there is a need to investigate ELLs from
a variety of minority ethnolinguistic groups.
A number of studies examined the effectiveness of specific instructional
approaches for students with special challenges. In most cases, these stu-
dents were identified as learning disabled or impaired on the basis of their
standing relative to district norms. It is possible that students identified
in this way face a heterogeneous group of challenges, including language
impairment, dyslexia, and learning disability. While these impairments
converge the longer students are in school, there are good theoretical and
practical reasons to believe that they are initially different forms of impair-
ment that, arguably, call for different forms of remediation (e.g., Bishop
and Snowling, 2004). For example, specific language impairment is not
the same as a general learning disability (see Leonard, 2000, for a review),
and appropriate intervention for the former is different from appropriate
intervention for the latter. Future researchers are encouraged to differen-
tiate special learning needs of ELLs and to identify learners with different
needs using appropriate selection criteria. Without more differentiation of
students with special challenges, our understanding of how to meet their
particular needs will remain sketchy.
When examined as a whole, this body of research suffers from the
“one-off” syndrome – that is, single studies by a researcher or team of
researchers on a specific pedagogical issue or approach in a specific school
and district. This style of research may reflect the pressures on university-
based researchers to “publish or perish” and/or the need to provide
answers quickly. Whatever the precise explanation, such an approach
leaves many unanswered questions about the reliability and generalizabil-
ity of reported results. Future research that entails long-term, sustained
P1: JZZ
efforts and multiple samples in different communities would contribute

significantly to our understanding of how and when these approaches
work. Also relevant to the issue of generalizability is the diversity of ways
in which variations of an approach, such as process-based approaches, are
operationalized. On the one hand, flexibility in the way an approach is or
can be operationalized is realistic and desirable so that the particular needs
and resources within particular classrooms can be addressed. On the other
hand, the lack of uniformity or coherence in the way approaches are imple-
mented compromises generalizability and replicability substantially. It is
incumbent on researchers to provide not only sufficient detail about the
implementation of the approach they are investigating to ensure that they
are studying what they say they are studying but also to provide informa-
tion about the reliability of their descriptions, which, as noted earlier, is
lacking in many cases.
language of instruction
An ongoing issue among theoreticians, researchers, and professional edu-
cators has been the benefits, or disadvantages, of providing instruction
through the medium of ELLs’ L1 (for more discussion of this issue, see
Chapter 5, this volume). A number of arguments for this (Cummins, 2000;
Thomas and Collier, 1997) and against it (Porter, 1990; Rossell and Baker,
1996) have been made. These will not be reviewed in detail here because
of space limitations. However, in short, proponents of L1 support argue
that L2 reading acquisition is facilitated if instruction is provided in a lan-
guage that students already know so that skills acquired in the L1 can
transfer to the acquisition of reading and writing in the L2. Opponents of
L1 support argue that it detracts from acquisition of the L2 because it gives
the learner less instructional time relevant to the L2 – the time-on-task
argument.
A lot of research has assessed the impact of instruction through the L1
versus instruction through English alone. The primary way in which this
issue has been addressed is by comparing the performance of ELLs who
have received instruction in the primary grades through the medium of
only English to that of ELLs who have received instruction in their L1
along with English instruction. Studies using these methods comprise the
majority of studies that were uncovered. Another way in which this issue
has been addressed is by examining the performance of ELLs in bilin-
gual education to that of native English speakers to ascertain if ELLs in
bilingual programs are more or less likely to achieve parity with native
English speakers (see Burnham-Massey and Piña, 1990). The corpus of
studies reviewed here is summarized in Table A.4.4. Chapter 5 provides
an extended review and critique of this and related research from the per-
spective of academic achievement.
P1: JZZ
Literacy 127
Before proceeding with a discussion of the results of this research, it is

useful to provide a methodological overview of them. The focus of this
work has been on elementary-level students since the case for bilingual
instruction or initial instruction through the students’ L1 applies primarily
to students who are receiving initial reading and writing instruction. The
choice of language of instruction to optimize educational outcomes is also
an issue for ELLs at the middle and high school levels, but we did not iden-
tify any research on these age groups. A number of studies have examined
the short-term impact of bilingual instruction on students’ literacy devel-
opment in Grades 1, 2, and 3 (Calderón et al., 1998; Carlisle and Beeman,
2000; Fulton-Scott and Calvin, 1983; Lindholm and Aclan, 1991; Saldate,
Mishra, and Medina, 1985); other studies have examined the medium-
term impact of bilingual instruction on students in the senior elementary
Grades 4, 5, and 6 (Burnham-Massey and Piña, 1990; Friedenberg, 1990;
Fulton-Scott and Calvin, 1983; Gersten and Woodward, 1995; Howard,
Christian, and Genesee, 2004; Lindholm and Aclan, 1991; Mortensen, 1984;
Ramirez, 1992); and some have examined the long-term impact on students
in Grades 7 and 8 (Bacon, Kidd, and Seaberg, 1982; Burnham-Massey and
Piña, 1990; Gersten and Woodward, 1995) and in one case in Grade 11
(Burnham-Massey and Piña, 1990). Examination of impact of bilingual
forms of education on student achievement in the early as well as later
elementary and middle grades is important because it highlights the impor-
tance of sustained, coherent instruction through the L1; we return to this
point later.
The majority of studies have employed standardized tests, including
the California Test of Basic Skills, Stanford Achievement Test, Iowa Test
of Basic Skills, Woodcock-Johnson, Metropolitan Achievement, SRA, and
Texas Assessment of Academic Skills. The use of standardized tests is an
important feature of these studies since it makes it possible to evaluate
practices related to language of instruction in the same way as in English-
only programs. Moreover, the population of students who contribute to
the norming of standardized tests is carefully selected to represent target
levels of achievement in school and, thus, use of such assessment instru-
ments ensures objectivity and generalizability that could be compromised
if only local instruments were used. Other outcome measures have also
been employed in single studies: GPAs (Fulton-Scott and Calvin, 1983) and
locally devised rubrics for scoring writing or oral language skills (Ferris
and Politzer, 1981; Howard, Christian, and Genesee, 2004; Kuball and Peck,
1997).
Overall, the studies within this corpus report similar results; namely,
that, in the long term, ELLs who receive some reading instruction in the L1
in the primary grades demonstrate the same or better performance in L2
reading as ELLs of similar linguistic and cultural background who have
received initial literacy and academic instruction in English only (Bacon,
P1: JZZ
Kidd, and Seaberg, 1982; Calderón, Hertz-Lazarowitz, and Slavin, 1998;

Carlisle and Beeman, 2000; Friedenberg, 1990; Fulton-Scott and Calvin,
1983; Gersten and Woodward, 1995; Lindholm and Aclan, 1991; Mortensen,
1984).
ELLs who receive initial instruction through the L1 do not always
demonstrate parity with comparison groups or test norms in the early
grades when L1 instruction is predominant. A number of studies have
found that it takes several years before parity is achieved. Gersten and
Woodward (1995) report advantages for ELLs in an English immersion
program in Grades 4 to 6 compared to ELLs in a transitional bilingual pro-
gram but no differences in Grade 7. Fulton-Scott and Calvin (1983) report
that there were few differences between ELLs in bilingual-multicultural
programs and ESL programs in the primary grades, but that the perfor-
mance of bilingually instructed students was superior to that of the com-
parison group by Grade 6. Calderón et al. (1998), Mortensen (1984), and
Saldate et al. (1985) also report that ELLs who received bilingual instruc-
tion demonstrated superior reading performance relative to comparison
groups, but again after participating in the program for some years.
Evidence of equal (or superior) reading achievement in English among
ELLs who received reading instruction in the L1 has also been reported
for ELLs with learning disabilities (Maldonado, 1994) and for ELLs of
Cherokee background (Bacon, Kidd, and Seaberg, 1982), two groups of
learners who are at added risk in school – risk due to their minority lan-
guage status and their learning handicap in the case of Maldonado’s learn-
ers and risk due to their minority indigenous cultural status in the case of
students in the Bacon et al. study. The Maldonado study is particularly note-
worthy because it entailed random assignment of ELLs with disabilities to
bilingual and English-only classrooms. More specifically, twenty students
with learning disabilities from one school were randomly assigned to either
an experimental group that received integrated bilingual special education
or a control group that received traditional special education in English.
The two groups were taught for three years in otherwise similar class-
rooms by similar teachers. The experimental group received instruction in
Spanish for all but 45 minutes per day during the first year and received
balanced language instruction (50 percent English; 50 percent Spanish) the
second year. English was the only language of instruction during the final
year. Instruction for the control group was conducted only in English. Per-
formance on the CTBS (pre- and post-test) showed that students in the
bilingual special education class were superior to students in the English-
only group at post-test. The students in the bilingual special education
program actually scored lower than the control group at pre-test, indicat-
ing that they had made even greater gains than the control group than
their post-test results alone would indicate. Replication of the Maldonado
study is warranted to address methodological weaknesses, in particular, to
P1: JZZ
Literacy 129
ensure that the only or primary instructional difference between the learn-
ing disabled group who received instruction through the L1 and those who
did not was the language of instruction.
The cumulative effects of participation in alternative educational pro-
grams for ELLs has been examined in a large-scale, longitudinal study by
Thomas and Collier (2002). Since that study is discussed in detail in Chap-
ter 5, it is not reviewed in detail here. Suffice to say that they report that
ELLs who were in mainstream English-only programs and had received
no special services scored lower by the end of high school than ELLs who
had participated in some form of bilingual education (i.e., early- or late-exit
bilingual or two-way immersion).
In a longitudinal study of the performance of Hispanic ELLs and English
L1 students in two-way immersion programs in Grades 3 to 5, Howard
et al. (2004) found that both ELL and English L1 students showed signif-
icant improvements in English reading and writing from Grades 3 to 5,
but that the ELLs performed significantly lower than the English L1 stu-
dents on both measures at all grade levels. ELLs in TWI programs received
initial literacy instruction in Spanish. The difference between the ELL and
English L1 groups could be due to a number of factors. First, the ELL
students were significantly more likely than the English L1 students to
come from lower socioeconomic backgrounds, as measured by parental
occupation or free/reduced lunch. Second, the ELL students may not have
been in the two-way immersion program long enough to reach parity with
their English-speaking peers. In support of this possibility, Collier (1987)
has reported that it can take from five to seven years for ELLs to achieve
grade-appropriate scores on standardized reading and language tests in
English (see also Cummins, 2000). ELLs have the triple challenge of acquir-
ing the societal language for both social and academic purposes, acquiring
new academic skills and knowledge, and adapting culturally to their new
environments, all at the same time. This calls for developmentally coher-
ent curricula that span several grades and for adaptations to assessment
programs that take into account ELLs’ long-term developmental trajecto-
ries. Longitudinal studies are critical if we are to ascertain and understand
the long-term results of particular instructional strategies or approaches, a
point we return to in Chapter 6.
Finally, Burnham-Massey and Piña (1990) compared the CTBS reading
and language subtest scores of ELLs who began instruction in Spanish
with those of English-only students longitudinally from Grades 1 to 12.
The ELLs’ percentile scores increased from Grades 1 to 5, at which time
they had reached the 46th percentile in reading and the 50th percentile
in language. Later comparisons in Grades 7 and 8 showed that the ELLs
scored at the same level as the English-only students; both scored around
the 50th percentile on language, but only at the 35th–41st percentile levels
on reading. The results were not compared statistically, and no explanation
P1: JZZ
is given for the drop in reading scores from Grade 5 to 7. The students were
subsequently compared in Grades 9 to 12 using GPAs and scores on the
High School Proficiency Test. The ELLs performed as well as the English-
only students at all grade levels on reading and writing, but again no statis-
tical analyses were carried out. Although these results must be interpreted
with caution for reasons that have been discussed, overall, they suggest
that the long-term language and reading development of ELLs who begin
elementary education in Spanish can be equivalent to that of English-only
students.
Taken together, the results of these studies provide little support for
the time-on-task argument against bilingual forms of education for ELLs.
Further evidence against this argument comes from a longitudinal inves-
tigation by Ramirez (1992) of three program alternatives for ELLs: struc-
tured immersion, early-exit bilingual, and late-exit bilingual. A subset of
Ramirez’s analyses compared the performance of ELLs in three versions
of late-exit bilingual education that varied with respect to the amount and
consistency of L1 instruction. Ramirez reports that ELLs “who were pro-
vided with a substantial and consistent primary language development
program learned . . . English language and English reading skills as fast or
faster than the norming population in the study” (Ramirez, 1992, p. 39).
In other words, more exposure to the L1 in school did not result in slower
L2 development, as would be predicted by the time-on-task argument.
Ramirez took considerable care to document the instructional practices of
the target programs in his investigation.
Additional evidence that questions the time-on-task hypothesis comes
from Ferris and Politzer (1981), who compared the writing skills of native
Mexican-born and U.S.-born ELLs. The former group had received their
first three years of education in Spanish in Mexico and their subsequent
education, until Grade 8, in English in the United States. The second group
was born in the United States and had received all their education in
English. Although the writing skills of the U.S.-born and educated group
were superior with respect to verb inflection, verb tense, and pronoun
agreement, there were no significant differences on many other measures of
writing: number of details, clarity, coherence, completeness; on frequency
counts of fused sentences, incorrect punctuation, article agreement, use of
possessives; or on T-unit analyses (average number of words per T-units,
average number of clauses per T-unit, average clause length).
Before leaving this section, a comment on random assignment might
be useful. Previous critiques of evaluations of bilingual education have
discounted this research on the grounds that most studies do not include
random assignment (Rossell and Baker, 1996). One of the primary argu-
ments for the use of random assignment is linked to the issue of gen-
eralizability. Random assignment of subjects to experimental conditions
permits researchers to rule out potentially confounding extraneous factors
P1: JZZ
Literacy 131
that might account for significant or nonsignificant effects. However, ran-

dom assignment is not always possible or appropriate when it comes to real
schools. First, it is difficult, for ethical and political reasons, for a school dis-
trict or ethics review board to enforce random assignment, without parental
permission, to such different educational options as English-only and bilin-
gual. Second, there is no realistic or legal way to ensure that participants
in bilingual programs would remain in programs following initial random
assignment; as a result, long-term participants in such programs could
no longer be said to be randomly assigned. In addition, since choice is
the hallmark of education in the United States and in current implementa-
tions of bilingual education, a randomly assigned group of students would
not be a valid reflection of the kinds of students and families who typ-
ically select such programs. However, convergent findings from studies
that have employed different assessment instruments and analytic tech-
niques, as well as different groups of students in different regions of the
country, is a realistic way of providing evidence for the generalizability
of the effectiveness of bilingual or any other form of instruction for ELLs
(see NRC Principle 5: Replicate and Generalize Across Studies, 2002). The
research reviewed here, arguably, provides this kind of evidence (see also
Chapter 5, this volume).
family and community

The studies reviewed in this section examine literacy development in rela-
tion to home-related factors, such as SES, language use at home, number of
books and literacy practices at home, parents’ values and aspirations, and
home-based factors, such as the impact of school-initiated interventions
in the home (e.g., the impact of using audio-books at home on students’
reading performance; Blum et al., 1995). Factors related to oral language
development in the home and its impact on phonological awareness, or
other reading-related skills, are discussed in more detail in Chapter 3. In
principle, one might also expect community-related factors to play a role
in the development of reading and writing skills in ELLs (e.g., the extent to
which written forms of language are evident and useful in the community).
In fact, we uncovered no studies of this nature.
Investigations of family and community factors are important because it
is often argued that ELLs are at-risk for reading failure or difficulty because
of their lack of exposure to or engagement in literacy outside school. This
assumption is itself subject to empirical verification. A thorough under-
standing of the language experiences of ELLs in the home and community
would be useful for developing school-based literacy activities that build
on these students’ total language experiences, especially those language
experiences that support literacy development. Moreover, evidence that
home- and community-based literacy activities can enhance ELLs’ literacy
P1: JZZ
development would be welcomed news because it could provide addi-

tional resources for promoting literacy development in school by drawing
on the assistance of parents and communities.
Two predominant variables in this research are socioeconomic status
and home-related literacy practices and resources. These two variables
are highly interrelated. SES was assessed in terms of eligibility for free
lunch. Eligibility for free lunch is linked to parental income which, in turn,
may be linked to parental occupation and education and, ultimately, to a
variety of literacy practices and resources in the home (e.g., literacy skills
of parents, their engagement in reading and writing on their own behalf
and with their children, and availability and use of books). It is important
to recognize that there may be a discrepancy between the current SES of
parents of ELLs and their education because immigrant parents may be
relegated to low-level jobs in their new communities despite high levels of
education and occupation in their communities of origin. We review these
two issues separately but recognize that they are likely concerned with the
same proximal family influences.
SES and Literacy Development

With respect to SES, Kennedy and Park (1994) and Reese et al. (2000) both
report significant correlations between SES and the standardized reading
test scores of middle-school ELLs. Kennedy and Park examined Grade 8
Asian and Mexican American ELLs while Reese et al. examined Grade 7
Latino students. Tompkins, Abramson, and Pritchard (1999) report that
ELLs with diverse L1 backgrounds attending relatively high SES schools
were more likely to use conventional English spelling patterns than ELLs in
less affluent schools. In contrast, Buriel and Cardoza (1988) report that SES
was unrelated to the standardized English reading and vocabulary scores
of Grade 9 first- and second-generation Mexican American ELLs living in
the Southwest. Personal aspirations were the most significant predictors
of these first- and second-generation students’ test results. SES made a
modest contribution to the vocabulary scores of third-generation students,
explaining an additional 3 percent of variance; SES did not account for
additional variance in the third-generation students’ reading scores. Ima
and Rumbaut failed to find a significant difference in SES between limited-
English proficient (LEP) and English proficient ELLs of Southeast Asian
background in regression analyses that examined the influence of a variety
of factors on school achievement. In line with the Ima and Rumbaut results,
Thomas and Collier (2002) also report that SES accounted for a relatively
small proportion of variance in the reading scores of ELLs in different
program options.
The discrepancy in these findings may relate to mitigating factors. SES
is itself not a causal variable but represents a number of other proximal
P1: JZZ
Literacy 133
variables that are causal in nature, such as reading practices and the avail-
ability of reading material at home. Thus, studies that include a constella-
tion of such proximal variables may yield different results with respect to
the influence of SES depending on which specific predictor variables are
included in the analyses and the validity and reliability of these other
variables. In other words, some multivariate studies may have elimi-
nated the statistical influence of SES on reading and/or writing scores,
thereby giving the impression that SES is unrelated to literacy develop-
ment. In fact, from developmental and pedagogical perspectives, the fam-
ily, individual, and sociocultural variables associated with SES are of more
significance than SES per se because they carry with them practical impli-
cations of some import. In contrast, SES alone implies little of action-
able consequence. This is evident as we turn to the next set of studies,
those that look at family-related literacy practices and their relationship to
literacy.
Family Literacy Practices

Studies of proficient and precocious ELL readers provide one source of evi-
dence for an association between home-related factors and literacy devel-
opment (Jackson and Wen-Hui, 1992; Pucci and Ulanoff, 1998). Pucci and
Ulanoff (1998) found that proficient ELL readers from minority-language
backgrounds had more books at home, enjoyed reading more, and felt they
were more proficient than less proficient ELL readers. The students in this
study were in Grade 4 and were Hispanic. In an interesting related study
of precocious ELL and precocious English L1 readers (in Kindergarten and
Grade 1), Jackson and Wen-Hui (1992) found that the two groups were
similar with respect to home-literacy experiences. More specifically, the
parents of both groups reported that their children read at home at least
two or three times a week; both groups were read to at home; both groups
had someone help them identify letter names, spell words, and under-
stand word meanings; and, in addition, both had attended some kind of
pre-school. However, the samples of these studies were quite small (twelve
in both cases), making generalizability difficult. More important, it is not
clear what the causal factors are in these studies – were these ELLs profi-
cient readers because of parental involvement in their children’s reading
or did these children’s precocity/proficiency in reading prompt parental
involvement, or both?
In a variation of these studies that have examined naturally occurring
home-based literacy practices, Blum et al. (1995) examined the effective-
ness of a school-initiated intervention that was conducted in the homes of
ELLs of diverse L1 backgrounds with no L1 reading ability. In the interven-
tion, audio recordings of stories were sent home with students to support
their reading at home in the absence of other sources of support. Blum et al.
P1: JZZ
report that the reading performance of these Grade 1 students benefited

from this initiative. However, their results are based on observation only
and their sample was small – only five students. This type of intervention
warrants more attention, with methodological enhancements, because of
its potential educational importance. The Blum et al. (1995) results along
with the Pucci and Ulanoff (1998) and Jackson and Wen-Hui (1992) results
have important implications for extending opportunities for literacy devel-
opment to include parents and home literacy training.
At the other end of the literacy spectrum, Hughes, Schumm, and Vaughn
(1999) examined the homes and parental literacy practices of ELLs with
learning disabilities (LD). More specifically, Hughes et al. compared the
homes of Grade 3 and Grade 4 Hispanic students with LD to those of sim-
ilar students with typical patterns of development with respect to types
of reading and writing activities, the desirability and feasibility of reading
activities, and facilitators and barriers to parents helping their children to
read. They found few differences between the homes of these two groups
of students – the most common activities in the homes of both groups
were reading to the children or having the children read to the parents;
both groups of parents valued home reading, but the parents of children
with LD found it more difficult to help their children read than did par-
ents with average or above-average readers; and both groups of parents
reported that the most common barriers to supporting their children’s
reading development at home were lack of communication with the school
and lack of knowledge of English. Evidence from this research on home-
related factors suggests that while parental and home-related factors may
be associated with learning/reading proficiency, they are less important in
explaining learning (i.e., reading) disability. Arguably, the latter is due to
endogenous limitations in children’s general cognitive or language-specific
capacity and, thus, may be relatively insensitive to variations in family
influences.
L1 Use at Home
A major issue concerning home/community factors and English literacy
development is the extent to which use of and/or proficiency in the L1
affects ELLs’ proficiency in English literacy. This question has been exam-
ined in a number of studies in which multiple independent variables have
been used to predict L2 literacy outcomes including, for example parental
levels of education and SES (as noted previously); educational aspirations,
expectations, and values; homework patterns; use of the L1 and L2 at
home; and immigration and medical background (Buriel and Cardoza,
1988; Duran and Weffer, 1992; Hansen, 1989; Ima and Rumbaut, 1989;
Kennedy and Park, 1994; Reese et al., 2000). Taken together, the results
from these studies suggest a complex multivariate relationship between
P1: JZZ
Literacy 135
L1/L2 use outside school and L2 literacy development in school. Impli-

cated in this complex relationship are generational status (e.g., whether
students are first-, second-, or third-generation immigrants; see Buriel
and Cardoza, 1988), the cultural backgrounds of students (e.g., whether
students are Mexican American or Asian American; see Kennedy and
Park, 1994), type of outcome measure (e.g., auditory vocabulary versus
text comprehension; see Hansen, 1989; and standardized test scores versus
course grades; see Kennedy and Park, 1994), and type of predictor measure
(e.g., oral language use versus literacy; see Reese et al., 2000).
Hansen (1989), also reviewed in Chapter 2, investigated the relationship
between gains in English reading (and auditory vocabulary) scores on the
one hand, and language use in the classroom, at home, and with peers
on the other hand, among Grade 2 and Grade 5 students from Spanish-
dominant homes in the San Francisco Bay area. They report that, after
prior levels of reading comprehension were accounted for, students’ use of
English at home and in the classroom accounted for significant portions of
variance in reading scores (18 and 9 percent, respectively). In other words,
ELLs who used more English at home and in the classroom scored higher
on tests of English reading than did ELLs who used English less often in
these settings.
Research by Kennedy and Park (1994) suggests that the significance of
L1/L2 use in the home depends on both students’ cultural background
and type of outcome measure. To be specific, they found that in the case
of Mexican American ELLs, English use in the home was not a significant
predictor of English course grades (r = 0.012) or of standardized English
test scores (r = −0.002). In the case of Asian American ELLs, English use
in the home was not related to course grades (r = 0.05) but was moder-
ately related to standardized test scores (r = 0.08∗ ). Kennedy and Park
note that the explanatory power of their final regression models, which
included other predictor variables, was quite small in all cases, accounting
for between 8 and 23 percent of the total variance in outcome measures,
depending on the measure and the background of the students. Home lan-
guage use was a much stronger correlate of standardized test results than
of course grades.
Buriel and Cardoza (1988) report that the influence of L1 use in the
home on L2 literacy development may also depend on the immigration
history of the student. The students in their study were first-, second-, or
third-generation Grade 9 students of Mexican descent living in the U.S.
Southwest. Regression analyses indicated that home language use was
not a significant predictor of English reading outcomes (as measured by
self-reported course grades and standardized reading tests) in first- and
second-generation students. In contrast, among third-generation students,
greater English dominance (based on self-reports of the extent to which
English was the first language, the current language of use in the home,
P1: JZZ
and usual individual language, among others) was associated with higher
scores on standardized English reading tests and vice versa for students
with less English dominance. Conversely, high Spanish-dominant third-
generation ELLs scored less well on English reading tests than low Spanish-
dominant students. In short, for these third-generation ELLs, there was a
positive correlation between extent of English use outside school and level
of English reading ability. At the same time, third-generation students with
greater literacy skills in Spanish scored higher on the English reading test
than did students with lower Spanish literacy skills. Home language use
did not predict English vocabulary scores among third-generation students
to any significant extent. In brief, high levels of both English language use at
home and Spanish literacy were correlated positively with level of English
literacy.
The dual significance of both oral English language use/proficiency and
Spanish literacy development is echoed in a study by Reese et al. (2000)
in which they monitored the English reading development of Spanish-
speaking students longitudinally from Kindergarten to Grade 7. Using path
analyses with multiple predictor variables and Grade 7 reading test scores
as the outcome variable, they report that the average number of years that
ELLs’ parents lived in the United States predicted students’ level of oral
proficiency in English (r = 0.32∗∗∗ ), which in turn predicted their English
reading scores in Grade 7 (r = 0.43∗∗∗ ). In other words, greater oral English
proficiency enhanced English literacy development. Duran and Weffer
(1992) and Ima and Rumbaut (1989) have similarly found that number of
years in the United States has a significant positive correlation with English
reading scores of Grade 10 Mexican American and Grades 7 to 12 South-
east Asian ELLs, respectively. Returning to Reese et al., they also found
that family literacy practices predicted early Spanish literacy (r = 0.22∗ ),
which in turn predicted Grade 7 English reading scores (r = 0.30∗∗ ). Inter-
estingly, their index of “family literacy practices” did not distinguish
between English and Spanish literacy and, thus, can be interpreted as an
index of literacy in either language. Thus, students who entered school with
some prior oral proficiency in English were advantaged in the long run for
the acquisition of English reading, as were ELLs who entered Kindergarten
with literacy skills in Spanish.
assessment
Assessment is undoubtedly one of the most critical aspects of education
for English language learners. It is implicated in virtually every aspect of
their education – from screening or admission, to identification of special
and individual needs that figure in instructional planning, to promotion or
retention. While there exists an extensive body of research on assessment
for native English speakers, this research is of dubious generalizability
P1: JZZ
Literacy 137
to ELLs for a variety of reasons. Most obviously, standardized or norm-

referenced tests are only valid for students on whom the test has been
normed, and use of such tests with other types of students can lead to
egregiously faulty results and decisions. For example, a standardized test
of mathematics or science administered in English to ELLs is just as much
about the student’s language proficiency as it is about his/her knowledge
of mathematics or science. Standardized tests that have been developed
for mainstream English-speaking students may contain cultural biases that
can result in underestimations of the competence of students from different
cultural and/or linguistic backgrounds (Cabello, 1984).
To be effective, education must be based on an accurate assessment
of students’ knowledge and skills; otherwise, students will be provided
instruction that is too advanced or not advanced enough, redundant, or
simply irrelevant. Despite its singular importance, research on assessment
of ELLs is dramatically lacking. Of the entire body of research reviewed
for this volume, only ten empirical studies related to assessment issues
were uncovered and retained. While these studies address issues of some
importance, their significance is weakened by conceptual fragmentation;
that is, each study looks at a different assessment issue. Thus, the findings
reported by these studies lack utility because we cannot ascertain their
reliability and generalizability.
The diversity of this research can best be illustrated by providing a brief
overview. McEvoy and Johnson (1989) examined the utility of using a test
of general intelligence (WPPSI) as a predictor of early reading scores among
Mexican American students. They found that, indeed, the WPPSI predicted
a significant amount of variance in reading scores when predicting from
age 5 to Grades 1 to 4. Jansky et al. (1989) failed to find good predictive
validity for a five-test screening battery for ELLs of Hispanic background
when initial screening occurred at Kindergarten or Grade 1 with follow-up
for five to six years. Accordingly, they argue for careful selection of screen-
ing tests that are fine-tuned to the specific needs and characteristics of the
students being tested. In contrast, Frontera and Horowizt (1995) report that
teachers can be a valuable and valid source of information concerning stu-
dents who are at risk for reading failure based on questionnaire responses.
Miramontes (1987) examined the miscues of good and disabled readers
whose first language was Spanish versus those whose first language was
English. She found that the miscues of students whose first language was
Spanish (whether they were good or disabled readers) adhered more con-
sistently to the expected graphic and sound cues represented in the text
than did those of students whose first language of reading was English.
She emphasizes the importance of assessing ELLs’ reading abilities in
both languages and also of examining reading processes, not just levels.
In another study, Miramontes (1990) makes the same claims as a result
of research on the miscues of Hispanic students with mixed language
P1: JZZ
dominance/proficiency. Umbel et al. (1992) also argue for assessment of

ELLs’ skills in both languages in order to arrive at valid assessments of
vocabulary skills. Cabello (1984) documents alternative forms of cultural
bias in tests used to assess ELLs – bias linked to lexico-syntactic, content
and concepts, social, and cognitive aspects of assessment. She recommends
the use of test items that are relatively free of bias when assessing students
with different linguistic and cultural backgrounds. Similarly, Garcia (1991)
found that the performance of Hispanic students on an English language
reading test was seriously underestimated because of their limited prior
knowledge of certain test topics and concludes that simply diversifying
the topics in a test is inadequate to overcome this bias. In a large survey of
5,472 students of Southeast Asian background, Ima and Rumbaut (1989)
argue that educators must consider the diversity of learners within this
group of ELLs if their educational efforts, including assessment activities,
are to be appropriate and effective.
Collectively, these studies address a number of important general issues.
However, it is difficult to provide recommendations from this corpus
because the research is so fragmented. Some observations that can be
derived from this corpus relate to importance of:
1. Assessment in both languages in order to arrive at complete and

valid assessments of ELLs’ abilities and difficulties (Miramontes,
1987; Umbel et al., 1992).
2. Multiple sources of information when assessing the learning needs
of ELLs (Frontera and Horowitz, 1995; McEvoy and Johnson, 1989).
3. Different kinds of information about ELLs and tests that are dynamic
and tailored to specific needs and characteristic of ELLs; in particular,
assessments should be free of cultural bias, sensitive to students’
relative proficiency in each language, and sensitive to developmental
patterns and to first-language reading (Miramontes, 1987).
4. Distinctive patterns of reading and language development among
ELL subgroups that can serve as valid points of reference for diag-
nosis – in other words, one should not assume that there are singular
or simple profiles that validly characterize the reading difficulties of
all ELLs who are below grade level (Goldstein, Harris, and Klein,
1993; Ima and Rumbaut, 1989; Merino, 1983; Miramontes, 1990).
There is clearly a need for much more research in this field if we are
to address these issues with empirical evidence. Additional key questions
deserve attention:
1. How can we best identify the strengths and weaknesses in ELLs’

L2 (and L1) reading skills and, in particular, those who are below
grade level in reading performance? In a related vein, do ELLs from
P1: JZZ
Literacy 139
different L1 backgrounds demonstrate the same patterns of difficulty

and strength?
2. Can the L2 reading skills of ELLs be calibrated in some way that
would permit appropriate placement of students in English reading
programs?
3. How can we distinguish ELLs who are suspected of impaired reading
from those who simply have incomplete mastery of the L2?
4. Can standardized English and content-area tests that are mandated
by state regulations be adapted for use with ELLs to make valid
inferences about their reading and academic progress relative to
mainstream students?
5. Is it possible to develop tests of English reading and language that
are not biased against ELLs in inappropriate ways?
conclusions
Studies of alternative approaches to reading and writing instruction indi-
cate that interactive approaches have much to recommend them. Most
evaluative studies on interactive approaches attest to their effectiveness.
Moreover, they are effective for students with typical as well as those with
impaired capacities for learning, although the evidence is limited and def-
initions of impairment are often overly general and lacking in precision.
We previously cautioned that the distinctions used to categorize studies
of instruction are not mutually exclusive. In fact, a number of studies con-
sist of combinations of approaches. Direct and interactive instruction was
frequently combined – a number of studies of classrooms that were catego-
rized as direct instruction included interactive components (e.g., Padron,
1992), and many interactive learning environments included direct-skills
instruction (Doherty et al., 2003; Goldenberg, 1992; Saunders and Golden-
berg, 1999). The evidence suggests that different forms of direct instruction
can be effective for teaching word-level and text-level language skills and
that it can be effective crosslinguistically. Our present understanding of
the effectiveness of direct instruction is limited by the relative paucity of
research on this approach.
The importance of direct instruction as a component of an overall plan
of instruction is indicated by the results of process-based approaches. The
effectiveness of process approaches is mixed at best, with some studies
reporting advantages for students who were in process-oriented literacy
classrooms, but many reported null advantages and even disadvantages.
Researchers who examined process approaches pointed out that simply
exposing students to literacy-rich learning environments is not sufficient to
promote acquisition of the specific skills that comprise reading and writing.
They argued that focused and explicit instruction in particular skills and
subskills is called for if ELLs are to become efficient and effective readers
P1: JZZ
and writers. Thus, the fundamental tenet of process-based approaches that

direct-skills instruction be avoided is the source of its weakness. Unless
process-based instruction is able to provide additional evidence to the con-
trary, the results to date do not recommend this approach.
The best recommendation to emerge from our review favors instruction
that combines interactive and direct approaches. Classrooms that combine
interactive with direct instruction provide instruction in specific reading
and writing skills within carefully designed interactive contexts, such as
Instructional Conversations. Interaction between learners and teachers, be
they adults or more competent students, is a mechanism through which
adaptation and accommodation of individual differences and preferences
can be accomplished. Carefully planned interactions in the classroom are
also both the medium for delivering appropriate instruction about liter-
acy and academic material and the message, insofar as the very language
that is used during interactive instruction embodies many key features of
language for literacy and broader academic purposes. Direct instruction
of specific skills ensures student mastery of literacy-related skills that are
often embedded in complex literacy or academic tasks. Presenting direct
instruction in interactive learning environments ensures that it is meaning-
ful, contextualized, and individualized. The choice of methods will depend
in large part on the objectives of instruction and learner characteristics. Cer-
tain methods, such as the keyword method, will be appropriate for vocabu-
lary development, while others, such as brainstorming, will be appropriate
for text comprehension and writing. At the same time, it is important to
point out that the emphasis on authentic, meaningful, and individualized
literacy activities that is the hallmark of process approaches is probably a
critical element of all literacy programs. It is a matter of balance and focus –
extant evidence suggests that an exclusive focus of the sort emphasized by
process approaches is probably not optimal.
A comprehensive and coherent plan for instruction calls for more than
specific techniques or methods, be they interactive or direct in nature.
Educators need comprehensive frameworks for planning and delivering
a whole curriculum. The Five Standards for Effective Pedagogy (Tharp
et al., 2000) provides such an overarching framework. “The Five Stan-
dards articulate both philosophical and pragmatic guidelines for effective
education. . . . they do not endorse a specific curriculum but, rather, estab-
lish principles for best teaching practices . . . for both majority and minority
students” (see CREDE, 2002, for more details). The standards emphasize
teaching and learning through interaction but also accommodate direct
instruction as appropriate. In light of the evident need to focus on spe-
cific components of oral language and reading and writing for academic
purposes, the SIOP framework developed by Echevarria, Vogt, and Short
(2000) deserves consideration at the same time. SIOP is of particular rel-
evance to the present discussion because it focuses on how language can
P1: JZZ
Literacy 141
be adapted to make academic content comprehensible and, vice versa,

how academic instruction can promote acquisition of language for aca-
demic purposes. It incorporates both interactive and direct instructional
approaches.
While direct and interactive instruction are generally effective, we do
not have a full understanding of the scope of their effectiveness. More
evidence with respect to the specific language-skills domains over which
each approach can be said to be effective is needed. There is considerable
evidence with respect to text-level reading skills but limited evidence
concerning vocabulary, spelling, phonological awareness and decoding,
and writing skills. By strategically focusing on different learner outcomes,
researchers could expand our understanding of the scope of effectiveness
of these approaches.
Research concerning the effectiveness of direct and interactive forms of
instruction alone or in combination at different grade/age levels would
provide valuable information about how to adapt these approaches to
learners as they progress developmentally. One might imagine that direct
instruction is particularly effective for acquisition of certain skills at certain
stages of development and less effective at other times for other kinds of
skills. Similarly, interactive instruction and particular forms of interactive
instruction alone or in combination with direct instruction might be par-
ticularly effective for teaching certain skills at particular stages of devel-
opment. The studies by Saunders and Goldenberg (1999) and Doherty
et al. (2003) provide research models for better understanding the effec-
tiveness of combined approaches to instruction and the components that
they comprise. More research that examines the relative merits of alter-
native approaches to instruction would provide educators with valuable
comparative data with which to decide among alternatives, as well.
Research with respect to language of instruction has focused on whether
use of ELLs’ L1 for initial schooling (and especially literacy instruction)
enhances, impedes, or has no effect on the development of reading and
writing skills in English. Collectively, this body of research indicates that
use of the L1 during the primary grades of schooling does not retard L2
literacy development. To the contrary, in some cases, ELLs who participate
in primary school programs that provide L1 support generally achieve the
same or superior levels of reading and writing skills as ELLs in English-
only programs by the end of elementary school or middle school. Thus,
contrary to the time-on-task hypothesis, ELLs can accommodate instruc-
tion in two languages without costs to their English-language develop-
ment. At the same time, ELLs who receive initial instruction in their L1
can achieve parity in English literacy with native English speakers and,
when they do, it usually takes several years of schooling. Participation in a
bilingual program does not lead to parity with native English speakers in
all cases, indicating that it is not sufficient to simply provide instruction in
P1: JZZ
ELLs’ L1 and English to achieve parity with native English speakers. Atten-
tion must also focus on type and quality of instruction. It is for this reason
that research on alternative instructional approaches for ELLs is so critical.
Unfortunately, there is insufficient research concerning additional instruc-
tional factors in bilingual classrooms to draw conclusions regarding the
joint effectiveness of language of instruction and instructional approach.
Moreover, the precise impact of use of the L1 for initial literacy instruc-
tion is not entirely clear since the sole independent variable in most of this
research is language of instruction with little control in most cases for other
possible mitigating factors.
Research on the role of family and community factors on L2 literacy
development reveals an interesting but complex picture. On the one hand,
ELLs who begin school with advanced levels of oral English language pro-
ficiency tend to achieve higher levels of English literacy than students who
lack proficiency in English and, conversely, ELLs who are Spanish domi-
nant upon school entry tend to achieve lower levels of proficiency in L2
literacy. On the other hand, ELLs with enriched literacy experiences during
the pre-school years in the home, whether in the L1 or the L2, and ELLs with
literacy skills in Spanish prior to school entry are likely to achieve higher
levels of proficiency in English L2 literacy than ELLs with less enriched
literacy experiences and less developed L1 literacy skills. The influence of
English language use/proficiency outside school on English literacy devel-
opment in school may also depend on the cultural background of students
as well as their parents’ length of residence in the United States. English
language proficiency upon entry to school appears to play a more impor-
tant role in English literacy development in school the longer the family
history of residence in the United States. Other factors, such as personal
aspirations, appear to be more important than L2 proficiency among first-
and second-generation immigrants. Oral English proficiency also appears
to be more important among Asian American than among Mexican Ameri-
can ELLs, although these results are statistically modest and are based on a
single study and, thus, deserve replication before substantive conclusions
can be drawn about the role of cultural background.
Taken in conjunction with the language-of-instruction findings, we see
that the acquisition of the home language and ESL is not a zero-sum game.
Proficiency in the home language does not have to result in reduced English
language skills and, conversely, the development of English language skills
does not have to entail loss of the home language. However, careful con-
sideration has to be given to what aspects of L1 and L2 development are
promoted in the pre-school years. The evidence reviewed in this chapter
and in Chapter 3 indicates that it is not sufficient to simply promote bilin-
gual proficiency in day-to-day oral language skills for ELLs to succeed fully
in English-language school environments. ELLs are more likely to succeed
at English-language literacy if they have had enriched literacy skills, either
P1: JZZ
Literacy 143
in the L1 or the L2, prior to school entry. Results reported in Chapter 3

indicate further that L1 oral language skills that are implicated in literacy
development are also likely to benefit L2 literacy development.
Parents and schools must play a greater role in promoting literacy in the
home, especially during the pre-school years so that ELLs are better pre-
pared for literacy instruction when they begin school. The same is probably
true for native English speakers who are at risk for literacy development.
Much more research is needed to examine family-based language inter-
ventions so that we have a better understanding of what works, with what
kinds of learners, and under what conditions. At present, our understand-
ing of the effect of home-based language interventions on English-language
literacy development in school is restricted. To date, ELLs of Hispanic back-
ground have been the primary focus of research attention, and studies of
non-Hispanic children have not disaggregated their results by language
and cultural background. There are good reasons to expect that specific
types of home-based interventions will be differentially effective for dif-
ferent learner groups (Steinberg, Dornbusch, and Brown, 1992). Future
research on these issues would fill these gaps in our knowledge.
Finally, with respect to assessment, it is difficult to draw precise or broad-
based conclusions because there is extremely little empirical research on
assessment issues related to ELLs, and the extant research is conceptu-
ally fragmented. A number of critical questions about assessment of ELLs
remains to be answered.
References
Avila, E., and Sadoski, M. 1996. Exploring new applications of the keyword method
to acquire English vocabulary. Language Learning 46(3), 379–95.
Bacon, H., Kidd, G., and Seaberg, J. 1982. The effectiveness of bilingual instruc-
tion with Cherokee Indian students. Journal of American Indian Education 21(2),
34–43.
Bermudez, A., and Prater, D. 1990. Using brainstorming and clustering with LEP
writers to develop elaboration skills. TESOL Quarterly, 24, 523–8.
Bishop, D. V. M., and Snowling, M. J. 2004. Developmental dyslexia and specific
language impairment: Same or different? Psychological Bulletin 130, 858–86.
Blum, I. H., Koskinen, P. S., Tennant, N., Parker, M. E., Straub, M., and Curry, C.
1995. Using audiotaped books to extend classroom literacy instruction into the
homes of second-language learners. Journal of Reading Behavior 27(4), 535–63.
Buriel, R., and Cardoza, D. 1998. Sociocultural correlates of achievement among
three generations of Mexican-American high school seniors. American Educa-
Burnham-Massey, L., and Piña, M. 1990. Effects of bilingual instruction on English
academic achievement of LEP students. Reading Improvement 27(2), 129–32.
Cabello, B. 1984. Cultural interference in reading comprehension: An alternative
explanation. Bilingual Review 11(1), 12–20.
P1: JZZ
Calderón, M., Hertz-Lazarowitz, R., and Slavin, R. 1998. Effects of bilingual coop-
erative integrated reading and composition on students making the transition
from Spanish to English reading. The Elementary School Journal 99(2), 153–65.
Carger, C. L. 1993. Louie comes to life: Pretend reading with second language
emergent readers. Language Arts 70(7), 542–7.
Carlisle, J. F., and Beeman, M. M. 2000. The effects of language of instruction on the
reading and writing achievement of first-grade Hispanic children. Scientific
Studies of Reading 4(4), 331–53.
Center for Research on Education, Diversity, & Excellence (CREDE) 2002.
The Five Standards for Effective Pedagogy. http://crede.ucsc.edu/standards/
standards.html.
Cohen, A. S., and Rodriquez, S. 1980. Experimental results that question the
Ramirez–Castaneda model for teaching reading to first-grade Mexican-
Americans. Reading Teacher 34(1), 12–18.
Collier, V. 1987. Age and rate of acquisition of second language for academic pur-
poses. TESOL Quarterly 21(4), 617–41.
Cummins, J. 1984. Bilingualism and special education: Issues in assessment and pedagogy.
San Diego: College Hill Press.
2000. Language, power, and pedagogy. Clevedon, UK: Multilingual Matters.
de la Luz Reyes, M. 1991. A process approach to literacy using dialogue journals
and literature logs with second language learners. Research in the Teaching of
English 25(3), 291–313.
Doherty, R. W., Hilberg, R. S., Pinal, A., and Tharp, R. 2003. Five standards and
student achievement. NABE Journal of Research and Practice Winter, 1–24.
Duran, B., and Weffer, R. 1992. Immigrants’ aspirations, high school process, and
academic outcomes. American Educational Research Journal 29(1), 163–81.
Echevarria, J. 1996. The effects of instructional conversations on the language and
concept development of Latino students with learning disabilities. The Bilin-
gual Research Journal 20(2), 339–63.
Echevarria, J., Short, D., and Powers, K. 2003. School reform and standards-based
education: How do teachers help English language learners? Technical Report. Santa
Cruz: Center for Research on Education, Diversity, & Excellence.
Echevarria, J., Vogt, M. E., and Short, D. J. 2000. Making content comprehensible for
English language learners: The SIOP Model. Needham Heights, MA: Allyn &
Bacon.
Fayden, T. 1997. What is the effect of shared reading on rural Native American and
Hispanic kindergarten children. Reading Improvement 34(1), 22–30.
Ferris, M., and Politzer, R. 1981. Effects of early and delayed second language
acquisition: English composition skills of Spanish speaking junior high school
students. TESOL Quarterly 15(3), 263–74.
Friedenberg, J. 1990. The effects of simultaneous bilingual reading instruction on
the development of English reading skills. Acta Paedologica 1(2), 117–24.
Frontera, L., and Horowitz, R. 1995. Reading study behaviors of fourth grade
Hispanics: Can teachers assess risk? Hispanic Journal of Behavioral Sciences 17(1),
100–20.
Fulton-Scott, M., and Calvin, A. 1983. Bilingual multi-cultural education vs. inte-
grated and non-integrated ESL instruction. NABE: The Journal for the National
Association for Bilingual Education 7(3), 12.
P1: JZZ
Literacy 145
Garcia, G. E. 1991. Factors influencing the English reading test performance of

Spanish-speaking Hispanic students. International Reading Association 26, 371–
92.
Gersten, R., and Woodward, J. 1995. A longitudinal study of transitional and immer-
sion bilingual education programs in one district. The Elementary School Journal
95(3), 223–39.
Goldenberg, C. 1992/93. Instructional conversations: Promoting comprehension
through discussion. The Reading Teacher 46, 316–326.
Goldstein, B., Harris, K., and Klein, M. 1993. Assessment of oral storytelling abil-
ities of Latino junior high school students with learning handicaps. Journal of
Learning Disabilities 26(2), 138–43.
Gomez, R., Parker, R., and Lara-Alecio, R. 1996. Process versus product writing
with limited English proficient students. The Bilingual Research Journal 20(2),
209–33.
Greene, J. P. 1998. A meta-analysis of the effectiveness of bilingual education research.
Claremont, CA: The Tomas Rivera Policy Institute.
Hansen, D. 1989. Locating learning: Second language gains and language use in
family, peer and classroom contexts. NABE: The Journal for the National Associ-
ation for Bilingual Education 13(2), 161–80.
Hernandez, J. S. 1991. Assisted performance in reading comprehension strategies
with non-English proficient students. The Journal of Educational Issues of Lan-
guage Minority Students 8, 91–112.
Hillocks, G., Jr. 1984. What works in teaching composition: A meta-analysis of
experimental treatment studies. American Journal of Education 93(1), 133–70.
Howard, E. R., Christian, D., and Genesee, F. 2004. The development of bilin-
gualism and biliteracy from grades 3 to 5: A summary of findings from
the CAL/CREDE study of two-way immersion education. CREDE Research
Report. Washington, DC, and Santa Cruz, CA: Center for Applied Linguistics
and Center for Research on Education, Diversity, & Excellence.
Hudelson, S. 1994. Literacy development of second language children. In F. Genesee
(ed.), Educating second language children (pp. 129–58). New York: Cambridge
University Press.
Hughes, M. T., Schumm, J. S., and Vaughn, S. 1999. Home literacy activities: Percep-
tions and practices of Hispanic parents of children with learning disabilities.
Learning Disability Quarterly 22(3), 224–35.
Ima, K., and Rumbaut, R. G. 1989. Southeast Asian refugees in American schools:
A comparison of fluent-English-proficient and limited-English-proficient stu-
dents. Topics in Language Disorders 9(3), 54–75.
Jackson, N. E., and Wen-Hui, L. 1992. Bilingual precocious readers of English. Roeper
Review 14(3), 115–19.
Jansky, J., Hoffman, M., Layton, J., and Sugar, F. 1989. Prediction: A six-year follow-
up. Annals of Dyslexia, 39, 227–46.
Jimenez, R. 1997. The strategic reading abilities and potential of five low-literacy
Latina/o readers in middle school. Reading Research Quarterly 32(3), 224–
43.
Kennedy, E., and Park, H. S. 1994. Home language as a predictor of academic
achievement: A comparative study of Mexican- and Asian-American youth.
The Journal of Research and Development in Education 27(3), 188–94.
P1: JZZ
Klingner, J., and Vaughn, S. 1996. Reciprocal teaching of reading comprehension

strategies for students with learning disabilities who use English as a second
language. The Elementary School Journal 96(3), 275–93.
2000. The helping behaviors of fifth graders while using collaborative strategic
reading during ESL content classes. TESOL Quarterly 31(1), 69–98.
Kramer, V. R., Schell, L. M., and Rubison, R. M. 1983. Auditory discrimination
training in English of Spanish-speaking children. Reading Improvement 20(3),
162–8.
Kuball, Y. E., and Peck, S. 1997. The effect of whole language instruction on the writ-
ing development of Spanish-speaking and English-speaking kindergartners.
Bilingual Research Journal 21(2–3), 213–31.
Kucer, S. 1992. Six bilingual Mexican-American students and their teacher’s inter-
pretations of cloze literacy lessons. The Elementary School Journal 92(5), 557–72.
1995. Guiding bilingual students “through” the literacy process. Language Arts
72(1), 20–9.
Kucer, S. B., and Silva, C. 1999. The English literacy development of bilingual
students within a transition whole-language curriculum. Bilingual Research
Journal 23(4), 345–71.
Leonard, L. B. 2000. Children with specific language impairment. Cambridge, MA: MIT
Press.
Li, D., and Nes, S. 2001. Using paired reading to help ESL students become fluent
and accurate readers. Reading Improvement 38(2), 50–61.
Lindholm, K., and Aclan, Z. 1991. Bilingual proficiency as a bridge to academic
173(2), 99–113.
Maldonado, J. A. 1994. Bilingual special education: Specific learning disabilities in
language and reading. Journal of Educational Issues of Language Minority Students
14, 127–47.
Martinez-Roldan, C., and Lopez-Robertson, J. 2000. Initiating literature circles in a
first grade bilingual classroom. The Reading Teacher 53(4), 270–81.
McEvoy, R., and Johnson, D. 1989. Comparison of an intelligence test and a screen-
ing battery as predictors of reading ability in low income, Mexican-American
children. Hispanic Journal of Behavioral Sciences 11(3), 274–82.
McLaughlin, B., August, D., Snow, C. E., Carlo, M., Dressler, C., White, C., Lively,
T., and Lippman, D. 2000. Vocabulary improvement and reading in English lan-
guage learners: An intervention study. Proceedings of a Research Symposium
on High Standards in Reading for Students from Diverse Language Groups.
Washington, DC: U.S. Department of Education.
Merino, B. J. 1983. Language development in normal and language handicapped
Spanish-speaking children. Hispanic Journal of Behavioral Sciences, 5(4), 379–
400.
Miramontes, O. 1987. Oral reading miscues of Hispanic students: Implications for
assessment of learning disabilities. Journal of Learning Disabilities 20(10), 627–32.
1990. A comparative study of English oral reading skills in differently schooled
groups of Hispanic students. Journal of Reading Behavior 22(4), 373–94.
Mortensen, E. 1984. Reading achievement of native Spanish-speaking elementary
students in bilingual vs. monolingual programs. Bilingual Review 11(3), 31–36.
P1: JZZ
Literacy 147
Ochs, E., and Schieffelin, B. B. 1984. Language acquisition and socialization: Three
developmental stories and their implications. In R. A. Schweder and R. A.
LeVine (eds.), Culture theory: Essays on mind, self, and emotion. New York: Harper
and Row.
Padron, Y. N. 1992. The effect of strategy instruction on bilingual students’ cognitive
strategy use in reading. Bilingual Research Journal 16(3–4), 35–51.
Porter, R. P. 1990. Forked tongue. New York: Basic Books.
Pucci, S., and Ulanoff, S. 1998. What predicts second language reading success? A
study of home and school variables. Review of Applied Linguistics 121–22, 1–18.
Ramirez, D. 1992. Executive summary. Bilingual Research Journal 16, 1–62.
Ramirez, S. Z. 1986. The effects of Suggestopedia in teaching English vocabulary to
Spanish-dominant Chicano third graders. The Elementary School Journal 86(3),
325–33.
Reese, L., Garnier, H., Gallimore, R., and Goldenberg, C. 2000. Longitudinal
analysis of the antecedents of emergent Spanish literacy and middle-school
Roser, N., Hoffman, J., and Farest, C. 1990. Language, literature, and at-risk chil-
dren. The Reading Teacher 43(8), 554–59.
Rossell, C. H., and Baker, K. 1996. The educational effectiveness of bilingual edu-
cation. Research in the Teaching of English 30, 7–74.
Rousseau, M. K., and Tam, B. K. Y. 1993. Increasing reading proficiency of language-
minority students with speech and language impairments. Education and Treat-
ment of Children 16(3), 254–71.
Saldate, M., IV, Mishra, S. P., and Medina, M., Jr. 1985. Bilingual instruction and
academic achievement: A longitudinal study. Journal of Instructional Psychology
12(1), 24–30.
Saunders, W., and Goldenberg, C. 1999. The effects of instructional conversations
and literature logs on the story comprehension and thematic understanding
of English proficient and limited English proficient students. The Elementary
School Journal 99(4), 277–301.
Schon, I., Hopkins, K. D., and Davis, W. A. 1982. The effects of books in Spanish
and free reading time on Hispanic students’ reading abilities and attitudes.
NABE: The Journal for the National Association for Bilingual Education 7(1), 13–20.
Schon, I., Hopkins, K. D., and Vojir, C. 1984. The effects of Spanish reading emphasis
on the English and Spanish reading abilities of Hispanic high school students.
The Bilingual Review, 11(1), 33–9.
Slavin, R. 1995. Cooperative learning: Theory, research and practice (2nd Ed.). Boston:
Allyn & Bacon.
Slavin, R. E., and Madden, N. A. 1999. Effects of bilingual and English as a second
language adaptations of Success for All on the reading achievement of students
acquiring English. Journal of Education for Students Placed at Risk 4, 393–416.
2001. One million children: Success for All. Thousand Oaks, CA: Corwin.
Snow, C. E., Burns, M. S., and Griffin, P. 1998. Preventing reading difficulties in young
children. Washington: National Academy of Sciences.
Steinberg, L., Dornbusch, S. M., and Brown, B. B. 1992. Ethnic differences in adoles-
cent achievement: An ecological perspective. American Psychologist 47, 723–9.
P1: JZZ
Syvanen, C. 1997. English as a second language students as cross-age tutors.

ORTESOL Journal 18, 33–41.
Tharp, R. 1997. From at-risk to excellence: Research, theory and principles of practice.
Research Report 1. Santa Cruz, CA: Center for Research on Education, Diver-
sity, & Excellence.
Tharp, R. G., Estrada, P., Dalton, S. S., and Yamauchi, L. 2000. Teaching transformed:
Achieving excellence, fairness, inclusion and harmony. Boulder, CO: Westview
Press.
Thomas, W. P., and Collier, V. P. 2002. A national study of school effectiveness for lan-
guage minority students’ long-term academic achievement. Santa Cruz, CA: Center
Thomas, W., Collier., V. P. 1997. School effectiveness for language minority students.
Washington: National Clearinghouse for Bilingual Education. (www.ncbe.
gwu.edu).
2002. A national study of school effectiveness for language minority students’ long term
academic achievement. Santa Cruz, CA: CREDE.
Tompkins, G. E., Abramson, S., and Pritchard, R. H. 1999. A multilingual per-
spective on spelling development in third and fourth grades. Multicultural
Education 6(3), 12–18.
Ulanoff, S. H., and Pucci, S. L. 1999. Learning words from books: The effects of Read
Aloud on second language vocabulary acquisition. Bilingual Research Journal
23(4), 409–22.
Umbel, V., Pearson, B., Fernandez, M., and Oller, D. 1992. Measuring bilingual
children’s receptive vocabulary. Child Development 63(4), 1012–20.
Vygotsky, L. 1962. Thought and language. Cambridge, MA: MIT Press.
Willig, A. C. 1985. A meta-analysis of selected studies of the effectiveness of bilin-
gual education. Review of Educational Research 55, 269–317.
P1: JZZ
0521859751c04
table a.4.1. Summary of Studies on Direct Instruction
Sample Instructional Outcome Measures

Authors Characteristics Grade Method(s) Research Design (partial listing) Results
CB989B/Genesee
Avila and Hispanic, low 5 Keyword Random – Cued recall of new – TR > CO immediate and
Sadoski achieving method in assignment to: vocabulary delayed recall of Eng
(1996) Spanish to – TR: keyword – Sentence words
enhance method completion using
acquisition of – CO: translation new vocabulary
0 521 85975 1
new English – n = 63 – Immediate and

vocabulary – Immediate and delayed recall of
delayed recall new vocabulary
(1 wk)
Bermudez Hispanic, urban, 3, 4 Brainstorming Random – Comprehension of – TR > CO elaboration
and Prater low income and clustering assignment to: stories – No sign differences on
(1990) to enhance rdg – TR: – Retention of story fluency, organization,
comp and wtg Brainstorming information comp, or retention
skills: Fluency, + clustering – Writing: Fluency,
elaboration, – CO: Discussion elaboration, and
and of questions organization
organization from basal measures
of ideas reader
November 12, 2005
– 45 minutes/ – n = 16 each
day for 6 days
15:37
(continued)
149
P1: JZZ
0521859751c04
150

Echevarria, ELLs in 1 East 6, 7, 8 Sheltered Convenience – Expository – TR > CO at post-test on
Short, and Coast (large, Instruction (SI) assignment to: writing: Lang total score, lang
CB989B/Genesee
Powers urban district) – TR: SI (n = 346) production, focus, production,

(2003) & 2 West Coast – CO: Regular support/ organization, mechanics
(large urban) instruction (n = elaboration, – TR = CO at post-test on
districts of the 94) organization, and focus and elaboration
United States, – Pre – post-testing mechanics (6-pt
0 521 85975 1
Hispanic, of both groups scale for each)

Asian/Pacific,
European
background,
diverse L1s
Hernandez Hispanic, LEP Summer Pre Modeling, Within group – Comp of Span – Stat. sign post-test
(1991) – Grade 7 monitoring design with stories improvement in Span
checklist, and pre – post- – Use of strategies reading comp
direct testing while reading Eng – All students exhibited
instruction to – Stats analysis of stories use of strategies during
promote use of gains in Span Eng reading – Evidence
reading comp read comp of transfer of skills
Strategies: – Narrative training from Span to
Question- description of Eng
November 12, 2005
generating, use of strategies

summarizing, in Eng
predicting –n=7
15:37
P1: JZZ
0521859751c04
Kramer, Hispanic, urban 1, 2, 3 Auditory Random assign to: – Discrimination of TR > CO: Pre- to post-test
Schell, and Kansas discrimination – TR = target and gains on trained and
Rubison training of 4 Discrimination non-target sounds untrained contrasts
(1983) contrasting training (n = 15)
sounds in – CO: No training
English (n = 15)
CB989B/Genesee
Kucer (1992) Hispanic, 3 – Cloze texts to Single subject – Videotapes and – 93% of students’
bilingual, teach design field notes of responses during cloze
working class, strategies for –n=6 lessons: Teacher’s were contextually
metropolitan comprehend- – Narrative understanding of appropriate
school ing unknown description of lessons – Students used
words in text post- – Literacy artifacts: cloze-based strategies as
0 521 85975 1
intervention Students’ use of taught but did not

performance strategies always explicitly
– Interviews: understand purpose of
Students’ and instruction in the same
teachers’ way as the teacher
retrospective intended
understanding of
lessons
(continued)
November 12, 2005
15:37
151
P1: JZZ
0521859751c04
152
CB989B/Genesee

McLaughlin Hispanic ELL 4, 5 – Vocab- Convenience –Test of knowledge Vocabulary:
et al. (2000) and Eng L1; enriched assignment to: of target words – After 1 yr: TR > CO
schools in lessons aimed – TR: – PPVT (ELL) on knowledge of
0 521 85975 1
Massachusetts, at improving Vocab-enriched – Test of polysemy target vocab; no sign diff

California, and vocabulary lessons production on PPVT or polysemy
Virginia and reading – CO: No – Test of – After 2 yrs: TR > CO
comp (12- intervention morphology (ELL) on vocab,
week – n = 150 in – Test of breadth of polysemy, morphology,
intervention) Grade 4 knowledge of and semantic
– n = 150 in vocab associations for both ELL
Grade 5 – Cloze test & Eng L1 gps; no sign
– Pre-post-testing (comprehension) diff on PPVT
– 2 yr participants = 1 yr
participants
– Gap btwn Gr 5 ELL and
Eng L1 reduced by 40%
following treatment
November 12, 2005
– ELL < Eng L1 on all

outcome measures in
yr 1
15:37
P1: JZZ
0521859751c04
Reading Comprehension
(cloze):
– After 1 yr: TR = CO
(ELL)
– After 2 yrs: TR > CO; 1
yr gp = 2 yr gp
CB989B/Genesee
– Gap btwn Grade 5 ELL

and Eng L1 students
reduced by 45%
Padron (1992) Hispanic, 3, 4, 5 Direct Random – Reading Strategy – Overall shift from use of
bilingual, low instruction assignment to: Questionnaire weak to use of strong
income, plus reciprocal – TR1: Reciprocal (RSQ) – Likert- reading strategies from
0 521 85975 1
suburban teaching or teaching of type scale of 14 Grade 3 to 4

Southwest Question- comp strategies strategies: 7 scales – TR gps used strong
Answer (n = 25) negatively related strategies (i.e.,
method to – TR2: Classify to reading summarizing,
promote use comp questions achievement and generating questions)
of: Question- according to 7 positively more than CO and used
generating, how they can be related to reading weak strategies (thinking
summarizing, answered achievement about something else
predicting, (n = 24) while reading, writing
and clarifying – CO1: Silent down every word in the
strategies reading + story) less than CO gp
answer – No gp differences on
questions other 10 strategies
November 12, 2005
(n = 21)
– CO2: Took RSQ
+ lesson in social
15:37
studies (n = 19)
153
– Pre- post-test
(continued)
P1: JZZ
0521859751c04
154

CB989B/Genesee
Rousseau Hispanic, with 11–12 yrs – Keyword and Single subject – Words read – Keyword was more
and Tam speech & old listening design with correctly effective than preview
(1993) language previewing alternating – Correct answers to method
impairment, methods to treatments and comprehension – Combined approach was
inner-city enhance reversal design questions more effective than
metropolitan reading: Oral (n = 5) single approach
0 521 85975 1
school decoding and – Numeric results

reading comp but no stat
analyses
Ulanoff and Hispanic, 3 – Concurrent Random – Test of vocabulary – Sign differences:
Pucci (1999) transitional Translation assignment to: – Immediate and Pre-Review > CO >
bilingual and Preview – – TR1 : Conc. delayed recall Translation gp
program Review Translation – Concurrent Translation
techniques to (n = 21) showed worst
promote – TR2 = Preview-
vocabulary Review (n = 23)
acquisition – CO = no
treatment
(n = 16)
November 12, 2005
– Pre-post testing
15:37
P1: JZZ
0521859751c04a
table a.4.2. Summary of Studies on Interactive Instruction

Blum et al. – ELLs with 1 Home-based Single subject – Weekly oral – All students showed
(1995) different L1s repeated design with readings: Oral improved reading
– Suburban reading with replications reading fluency, fluency and accuracy
CB989B/Genesee
school in audio-model –n=5 self-monitoring over time

Washington – Narrative behavior and – Enhanced frequency of
metropolitan description motivation and involvement with
area – Observation reading at home
– No L1 reading surveys: Letter
ability identification,
0 521 85975 1
word recognition,
other oral reading
behaviors
Calderón, Hispanic; 2, 3 Bilingual Btwn (matched) – Grade 2: Texas Grade 2:
Hertz- Spanish- Cooperative, groups: Assessment of – TR > CO on rdg and
Lazarowitz, dominant Integrated – TR: (BCIRC) Academic Skills wtg TAAS
and Slavin ELLs Reading and n = 129 (Sp TAAS): Grade 3:
(1998) Composition – CO: Traditional reading and – TR > CO Eng reading
(BCIRC) instruction writing NAPT
n = 93 – Grade 3: – TR = CO on Eng
Norm-referenced language NAPT
Assessment – 2 yrs of TR > 1 yr of
Program for TR > CO
November 12, 2005
Texas (Eng – TR > CO on Grade 3 exit

NAPT): reading criteria to all-English
and writing program
16:36
155
(continued)
P1: JZZ
0521859751c04a
156
table a.4.2. (continued)
CB989B/Genesee

Cohen and Hispanic, low 1 Group-oriented Random – CTBS: reading – TR1 > TR2
Rodriquez SES, urban (interactive) assignment to: comprehension
(1980) vs. direct, – TR1: HIL subtest
0 521 85975 1
high intensity (approx. n = 75)

(HIL) reading – TR2: interactive
instruction group-oriented
instruction
(approx. n = 75)
Doherty et al. – Largely 3, 4, 5 Comprehensive Regression – SAT-9: – Higher use of Five
(2003) Hispanic program analyses to Comprehension, Standards reliably
– Low income using the Five examine results language, predicted gains on SAT:
– Low- Standards for of students in reading, spelling, Comp, reading, spelling,
performing Effective programs with vocabulary, and vocab
rural school in Pedagogy; high vs. low overall NCE
California emphasis on conformity to
interactive model
November 12, 2005
pedagogy – n = 266
thru IC students; 15
teachers
16:36
P1: JZZ
0521859751c04a
Echevarria – Hispanic ELLs 7, 8, 9 Instructional Within subject, – Amount of – IC > BR on academic

(1996) with learning Conversations alternating academic discourse, total number
disabilities, (IC) treatment discourse of utterances, and
metropolitan design with 5 IC – Utterances self-initiated utterances
Los Angeles and 5 BR lessons (number of): total, (scr. and non-scr)
– Assessed self-initiated – IC > BR on concept
fidelity of scripted (scr) & understanding
CB989B/Genesee
implementation self-initiated non- – IC = BR on narrative

– n = 5 (n = 3 for scripted (non-scr) structure and
some analyses) – Narratives: propositions
Structure, – IC = BR on literal recall
number of – High corr btwn IC
propositions implementation and
0 521 85975 1
– Thematic concept: student participation

level of (r = 0.98)
conceptualization
of theme
– Literal recall
– Student
participation
Fayden – Native K Shared reading Within subject – Clay’s Reading – Significant improvement
(1997) American and design with Strategies on both tests from pre- to
Hispanic, rural pre-post-testing – Sand Test: post-test
– Low SES – n = 24 Elementary
– Limited knowledge of
exposure to printed material
November 12, 2005
books
(continued)
16:36
157
P1: JZZ
0521859751c04a
158

Goldenberg – Hispanic ELLs 4 Instructional Btwn group: – Literal comp: – TR = CO on literal
(1992) in transition Conversations – TR: IC lesson on responses to 10 comprehension (75% –
CB989B/Genesee
from bilingual reading about open-ended, short- 80%)

to English-only friendship answer questions – TR > CO on essay: e.g.,
instruction – CO: Basal lesson – Conceptual reference to complex
on reading understanding: ideas – TR (62%) vs CO
about friendship essay on (14%)
friendship
0 521 85975 1
Klingner and Hispanic, urban, 7, 8 Reciprocal Random – Woodcock – Sign pre- post-test
Vaughn middle class, teaching with assignment to: Johnson: improvement on rdg
(1996) with learning cross-age – TR1: Tutorial gp letter-word comp and strategy use
disabilities tutoring or (n – 13) identification, for both gps
cooperative – TR2: passage – No sign differences
learning to Cooperative gp comprehension btwn coop and tutor gps
promote (n = 13) – Woodcock (both interactive)
reading com- – Pre- post-testing Language – More improvement
prehension Proficiency Battery during teacher-assisted
(Span) phase than coop or
– LAS (Eng and tutoring phase
Span)–
Gates-MacGinitie
November 12, 2005
Reading
comprehension
– Interview: reading
16:36
strategies
P1: JZZ
0521859751c04a
Klingner and Hispanic, 5 Collaborative Within subject – Vocabulary test – Sign pre- post-test gains
Vaughn metropolitan strategic design with pre- (researcher- on vocab
(2000) school, reading post-testing made); stat – Students spent lots of
Southeast technique and – n = 37 analysis time engaged in
United States its effects on – Stats used for – Use of reading academic-related
vocabulary vocab results strategies and strategic discussion and
CB989B/Genesee
comprehen- only helping behaviors assisted one another in

sion and on during understanding word
the use of audiotaped meanings, getting the
strategies to cooperative main idea, asking and
help one learning group answering questions,
another sessions and relating old
0 521 85975 1
understand (narrative knowledge to new

vocab analysis) knowledge
Kucer (1992) Hispanic, 3 – Cloze texts to Single-subject – Videotapes and – 93% of students’
bilingual, teach design field notes of responses during cloze
working class, strategies for –n=6 lessons: teacher’s were contextually
metropolitan comprehend- – Narrative understanding of appropriate
school ing unknown description of lessons – Students used
words in text post- – Literacy artifacts: cloze-based strategies as
intervention students’ use of taught, but did not
performance strategies always explicitly
– Interviews: understand purpose of
students’ and instruction in the same
teachers’ way as the teacher
November 12, 2005
retrospective intended
understanding of
lessons
16:36
159
(continued)
P1: JZZ
0521859751c04a
160

Li and Nes – Chinese L1 1, 2, 3 Audiotaped Single-subject – Oral reading rate – Fluency and accuracy
(2001) with ESL paired design with and fluency increased over 8 mths of
CB989B/Genesee
instruction reading pre-post- intervention

– Low Eng observation
reading levels –n=4
– West Texas, – Narrative
urban description
Martinez- – Latino, Spanish 2 Literature circles Narrative – Children’s – Young bilingual children
0 521 85975 1
Roldan dominant and description of engagement in can engage in rich

and Lopez- English behavior of and benefits from discussion of children’s
Robertson dominant Span dominant literature-circle literature
(2000) – Tucson, (n = 8) and Eng discussions
Arizona dominant
(n = 14)
students during
literature circles
Padron (1992) – Hispanic, 3, 4, 5 Reciprocal Random – Reading Strategy – Shift from use of weak to
bilingual, low teaching and assignment to: Questionnaire use of strong reading
income direct – TR (2 gps) strategies from Grade 3
– Suburban, SW instruction in – CO (2 gps) to 4
use of – See Table A.4.1 – TR gps used more
November 12, 2005
cognitive for details strong strategies and

reading – Pre–post-testing CO gps used more weak
strategies strategies
16:36
P1: JZZ
0521859751c04a
Saunders and – Hispanic (Hisp) 4, 5 Instructional Btwn group: ELL – Factual – Factual comp: (LL + IC)
Goldenberg ELLs Conversations and Eng comprehension of = IC > LL = CO for
(1999) – English- (IC) and proficient written stories both Eng + Hisp gps
proficient Literature students – Interpretive comp – Interpretive comp:
students Logs (LL) assigned to: of stories (LL + IC) = IC > LL =
– Low- – TR1: IC alone – Understanding of CO for both gps
performing – TR2: LL alone story theme: – Theme explanation:
CB989B/Genesee
school; low – TR3: IC + LL explanation and – Eng prof students: (LL +

income – CO: Regular exemplification IC) = LL = IC > CO
instruction – ELL: (LL + IC) > IC =
LL = CO
– Theme exemplification:
0 521 85975 1
(LL + IC) > IC = LL =

CO for all gps
Syvanen Hispanic, 4, 5 Cross-age Within-subject – MMCPC scale: – No sign diff btwn
(1997) Spanish- tutoring; design with pre- Perception of sample’s rdg score gains
dominant, ESL 19-week post-testing control of and gains of sample of
students intervention – n = 16 cognitive/social district students
domains by tutors – Pre- to post-test gains in
– District reading self-control over
achievement test tutoring process, and
– Survey of attitudes toward
attitudes toward reading
school – No increase in interest in
school and reading
achievement
November 12, 2005
16:36
161
P1: JZZ
table a.4.3. Summary of Studies on Process Instruction

0521859751c04a
162
Authors Characteristics Grade Method(s) Research Design (partial list) Results
Carger (1993) Hispanic; K Book-sharing and Single-subject – Engagement in – Children grew in ability to
emergent pretend design reading convey meaning with
readers reading in –n=3 – Word and idea emotion and confidence
groups – Narrative counts during – Comfort level with
CB989B/Genesee
description audiotaped reading also increased

pretend reading
sessions
de la Luz Hispanic, Span 6 Use of dialogue Single-subject Analysis of literature – ELLs attempted to write in
Reyes L1, bilingual journals and design logs: Eng before they had
(1991) students, low to literature logs – n = 10 – Topics and themes, complete control over
0 521 85975 1
middle SES to promote – Narrative sensitivity to language

writing description audience – Development of complex
– Self-concept and ideas and construction of
attitudes meaning suffered
– Language use
(Span vs. Eng)
– Length and quality
of writing
– Spelling and
grammar
Gomez, – Hispanic ELLs, Summer, Free writing vs. Random Analysis of writing TR > CO on analytic and
Parker, and low-achieving, Pre-6 structured assignment to: skills: holistic scores, but few
Lara-Alecio southeast Texas writing – TR: Structured – Micro-level stat sign differences
(1996) gp. (3 classes) indicators
November 12, 2005
– CO: Free writing – Analytic ratings

(5 classes) – Holistic ratings
– n = 48
16:36
P1: JZZ
0521859751c04a
Kuball and – Hispanic ELLs K Whole language Btwn group design: – Writing Measures: Writing of ELLs and Eng L1
Peck (1997) and Eng L1s, – Hispanic (n = 8) 8 stages of students improved over 1
low income, Los – English L1 (n = 8) grapho-phonemic year: # students who
Angeles, urban – Pre-post analysis skill development viewed themselves as
– Narrative 4 stages of writers, ratings of
description composition skills composition skills,
development grapho-phonemic skills
CB989B/Genesee
– Student
Questionnaire:
self-concept as
writers
Kucer (1995) Hispanic, 3 Use of wall charts Single-subject – Observations of – Exposing students
0 521 85975 1
bilingual, to promote use design students’ use of explicitly to alternative

literate in of reading, – n = unstated strategies during strategies through wall
Spanish reader – Narrative reading and charts did not ensure they
response, description writing would apply them during
spelling, and rdg and wtg activities
writing – Motivation is also
strategies in important
whole-language
classroom
(continued)
November 12, 2005
16:36
163
P1: JZZ
0521859751c04a
164

Authors Characteristics Grade Method(s) Research Design (partial list) Results
Kucer and – Hispanic, 3 Transitional Within subject – Oral reading and – Significant gains in
Silva (1999) bilingual, whole design with pre- story retelling reading, based on miscue
CB989B/Genesee
working class, language post-testing – Story writing and analysis

metropolitan bilingual class – n = 26 spelling – No gains in writing
area – Field notes and
interviews
Martinez- – Latino, Spanish 2 Literature circles Narrative – Children’s – Young bilingual children
0 521 85975 1
Roldan and dominant, and description of engagement in and can engage in rich
Lopez- English behavior of Span benefits from discussion of children’s
Robertson dominant dominant (n = 8) literature circle literature
(2000) – Tucson, Arizona and Eng discussions
dominant (n =
14) students
during literature
circles
Roser, – Hispanic ELLs, K, 1, 2 Shared literature Convenience – Student/teacher – TR had greater 1-yr gains
Hoffman, low SES, assignment to: reports on their than CO (in percentile
and Farest Brownsville, – TR = Shared reactions to points):
(1990) Texas reading (6 program – Language arts: TR = 10.5
schools) – California Test of pts; CO = 6.3 pts
– CO = 6 control Basic Skills: – Reading: TR = 14 pts;
schools in district Standardized
November 12, 2005
CO = 3 pts
– Narrative reading test for – Equivalence btwn TR and
description 2nd grade (no stat CO schools was not
analysis) determined
16:36
P1: JZZ
0521859751c04a
Schon, – Hispanic, low 2, 3, 4 Free reading time Convenience – Inter-America – TR = CO on English

Hopkins, SES, Tempe, with variety of assignment to: Reading Test: reading comp,
and Davis Arizona L1 (Span) – TR: Free reading reading comp, vocabulary + speed
(1982) books with variety of vocab, speed – TR > CO on Span reading
Span books subtests – Attitudes of TR students
(n = 49) – Researcher- improved
– CO: Usual developed survey:
CB989B/Genesee
method of self-concept and

instruction – attitudes toward
emphasis on Eng reading
reading (n = 44)
Schon, Study 1: 9–12 Free reading time Study 1: Study 1: Study 1:
0 521 85975 1
Hopkins, – Hispanic, low with high- For Eng results: – Metropolitan TR = CO on Eng reading
and Vojir SES, school in interest books – TR (n = 64): Reading Test comprehension
(1984) affluent area of in Span L1 Enriched reading – Measures of Study 2:
Tempe, Arizona materials and attitudes toward TR = CO on Eng reading
Study 2: free reading time reading and self comprehension
– Hispanic, low – CO (n = 47): Study 2:
SES, school in Other remedial – Nelson Reading
middle-income classes Test
area For affective – Measures of
results: attitudes toward
TR (n = 44) vs. CO reading and self
(n = 44)
Study 2:
– For Eng results:
– TR (n = 18) vs.
November 12, 2005
CO (n = 21)
– For affective
results:
16:36
165
TR (n = 33) vs.
CO (28)
P1: JZZ
table a.4.4. Summary of Studies on Language of Instruction

0521859751c04a
166
Authors Characteristics Grade Research Design (partial list) Results
Bacon, Kidd, and Cherokee ELLs 8 Convenience assignment SRA Achievement Tests: TR > CO
Seaberg (1982) to: Reading
– TR: bilingual program
in Grades 1–5 (n = 35)
– CO: English-only
CB989B/Genesee
instruction (n = 18)
Burnham-Massey Hispanic, former 1–5, 7–12 Convenience assignment – CTBS: Reading and – TR showed steady
and Piña (1990) LEP students, to: language subtests increase in CTBS
and English L1 – TR: Former LEP who (Gr. 1–5, 7–8) reading and language
students had had reading – Grade point averages scores from Gr. 1–5
instruction in Span (Gr. 9–12) – Grade 5: TR scored at
0 521 85975 1
(n = 117) – High School 50%ile in language &

– CO: Eng L1 students Proficiency Tests 46%ile in reading
(n = 492) (Gr. 9–12) – Grade 7/8: TR = CO,
– Longitudinal scored at 45–56%ile on
– No statistical language but 35–41%ile
comparisons on reading
– Grades 9–12: TR = or >
CO (percentage
students passing test)
Calderon, Hispanic, Spanish 2, 3 Btwn (matched) groups: – Texas Assessment of – TR > CO English
Hertz-Lazarowitz, dominant, low – TR: Bilingual Academic Skills reading
and Slavin (1998) SES, El Paso, cooperative, integrated (Spanish) – TR = CO on English
Texas reading and – English language
November 12, 2005
composition (n = 129) norm-referenced – 2-yr TR > 1-yr TR > CO

Assessment Program
for Texas
16:36
P1: JZZ
0521859751c04a
– CO: Cooperative – TR > CO on Grade 3

learning without exit criteria to
bilingual integrated all-English program
approach (n = 93)
Carlisle and Hispanic, Span at 1 Convenience assignment – Woodcock-Johnson – TR = CO on Eng
Beeman (2000) home (60%), Eng to: (Span, Eng): reading, reading
CB989B/Genesee
at home (28%), – TR (n = 19): Spanish listening, writing – TR > CO on Span

both Eng and instruction reading
Span at home – CO (n = 17): English – Instruction in Span
(12%), small town instruction enhanced Span reading
in California – No effect of instruction
in Eng on
0 521 85975 1
Eng reading
Ferris and Politzer – Hispanic, born in 7, 8 Convenience assignment – Multiple measures of – CO > TR on verb
(1981) Mexico or in to: writing skills: holistic, inflection, verb tense,
United States – TR (n = 30): Born and T-units, error analyses and pro agreement
– TR < CO on SES educated in Mexico til – TR = CO on all other
– TR > CO on Grade 3 measures
Spanish lang – CO (n = 30): Born and
educated in United
States
Friedenberg (1990) Hispanic, former 3, 4 Convenience assignment Stanford Achievement TR > CO
ESL students, to: Test: reading
Southern Florida – TR (n = 249): reading
instruction in Eng and
November 12, 2005
Span
– CO (n = 53): reading
instruction in English
16:36
only
167
(continued)
P1: JZZ
0521859751c04a
168

CB989B/Genesee
Fulton-Scott and Hispanic, 3 1, 6 Convenience assignment – gpa in reading and – TR > CO1 and CO2:
Calvin (1983) Southern to: language gpa, CTBS
California urban – TR: – CTBS
schools Bilingual-multicultural
– CO1: Integrated ESL
– CO2: Non-integrated
0 521 85975 1
ESL (n = approx. 10
students per grade
level at each school)
– Total n = 59, gp n’s not
specified
Gersten and Hispanic, TR and 4, 5, 6, 7 Convenience assignment ITBS: Language, reading, – TR > CO in Grades 4, 5,
Woodward (1995) CO equivalent on to: and vocabulary and 6 on language test
SES, low level of – TR: Bilingual – TR = CO in Grade 7 on
English on entry, immersion (n = 111, 5 language
10 schools in El schools) – TR > CO in Grade 4 – 5
Paso, Texas – CO: Transitional on rdg and gr 5 on
bilingual (n = 117, 5 vocab
schools) – TR = CO in Grade 7 on
November 12, 2005
– TR and CO schools rdg and vocab

similar characteristics – Both TR and CO scored
– Longitudinal low
16:36
P1: JZZ
0521859751c04a
Howard, Christian, – Hispanic, 3, 4, 5 Btwn group design: – Cloze reading test in – Span and Eng students
and Genesee (in primarily low – Span (n = 153–162) Span and Eng showed improvement
press) SES – Eng (n = 148–167) – Narrative writing in Span and Eng
– English L1, samples collected 9 reading and writing
primarily middle – n varies by analysis times over 3 years from Grades 3 to 5 –
class – Longitudinal design Exhibited similar
– 11 two-way growth curves
CB989B/Genesee
immersion sites – Eng gp > Span gp on

in United States Eng reading and
writing measures at all
grade levels (sign.)
Lindholm and – Spanish L1 1–4 Two-way Spanish CTBS: Reading subtest – Spanish students
0 521 85975 1
Aclan (1991) – English L1 immersion group approached grade level

students compared to test norms in English reading by
– 2 schools in – total n = 249 Grade 4
Northern – High proficient
California bilinguals >
medium/low proficient
bilinguals
Maldonado (1994) Hispanic, learning 2, 3 Random assignment to: CTBS: Reading and – CO > TR at pre-test
disabled, inner – TR: Integrated bilingual language – TR > CO at post-test
city, middle and special education
low SES, (n = 10) vs.
Houston, Texas – CO: English-only
special education –
(n = 10)
November 12, 2005
– Pre-post-analysis
(continued)
16:36
169
P1: JZZ

0521859751c04a
170
Mortensen (1984) Hispanic 4, 5, 6 Convenience assignment Wisconsin Design Tests – TR = CO word attack
to: for Reading Skills skills
– TR: Bilingual Development – TR > CO
instruction (n = 65) comprehension skills
– CO: Eng only (n = 55)
Ramirez (1992) (See Hispanic ELLs K–6 Btwn (matched) group: CTBS (Eng): Language – TRI = TR2 on lang and
CB989B/Genesee
Chapter 5 for longitu- – TR1: English and reading subtests reading tests by Grade
greater details of dinal immersion 6
this study) – TR2: Early-exit – Students in all 3
bilingual programs increased
– TR3: Late-exit bilingual Eng lang and reading
0 521 85975 1
– Total n = 2,000 skills at the same rate

from K to Grade 1 and
from Grade 1 to Grade
3
– Students in late-exit
who were given
substantial and
consistent L1
instruction increased
their rate of Eng lang
and rdg skills as fast as
or faster than norming
gp
Saldate, Mishra, Hispanic, low SES, 1–3 Convenience assignment – MAT: Eng or Span – TR = CO in Grade 2
November 12, 2005
and Medina Arizona to: (Grade 2) – TR > CO in Grade 3

(1985) – TR: Bilingual Program – WRAT: Reading, Eng,
(n = 30) or Sp. (Grade 3)
16:36
– CO: Eng only program

(n = 31)
P1: JZZ
0521859751c04a
table a.4.5. Summary of Studies of Family Influences

CB989B/Genesee

Blum et al. (1995) ELLs with diverse 1 Single-subject, reversal Weekly oral reading: – Use of audio-recordings
L1s, no L1 reading design ABA with – Oral reading fluency, at home during reading
ability, suburban multiple baselines self–monitoring enhanced reading
school in –n=5 behavior, and performance
metropolitan – Narrative description motivation – Students expressed
0 521 85975 1
Washington Marie Clay observation preference for audiotape

survey: method
– Letter identification,
word recognition,
hearing and recording
sounds in words, and
oral reading behavior
Buriel and Cardoza 1st, 2nd, and 3rd 9+ Within–subject design – Standardized test of – L1 home use was
(1988) generation with regression analysis reading and vocabulary unrelated to reading in
Mexican American Between-subject: 1st vs. 1st generation; modest
students, U.S. 2nd vs. 3rd generation relationship in 2nd
Southwest – n = 103–37 depending on generation; moderate
analysis and mixed relationship
with 3rd generation
November 12, 2005
– SES unrelated to reading

in 1st and 2nd generation
and only moderately
16:36
related in 3rd generation
171
(continued)
P1: JZZ

0521859751c04a
172
Duran and Weffer Top 25% of high 10 Within-subject design with – American College – Yrs in United States had
(1992) school grads regression analyses Testing: reading subtest significant effect on
(talented), – n = 157 – GPA pre-high school reading
Mexican – Family values did not
American, large affect reading ach.
CB989B/Genesee
midwestern city directly but did affect

ach. through
classroom-related
behaviors
Hansen (1989) Mexican American, 2, 5 Within-subject design with – Stanford Diagnostic – Auditory vocabulary
greater San regression analyses Reading Test: grew during summer, but
0 521 85975 1
Francisco, inner – n = 117 comprehension & text comprehension did

city, residential auditory vocabulary not
urban, and subtests – Peer language use during
suburban settings summer was related to
auditory vocabulary
gains
Hughes, Schumm, Hispanic parents of 3–5 Btwn group: – Types of reading & – Few differences between
and Vaughn (1999) learning disabled – parents of LD children writing activities at parent groups
and average to (n = 40) home – Most common activity
high average – Parents of children with – Desirability & feasibility was reading to children
readers average or of activities or having child read to
above-average reading – Facilitators & barriers to parents
skills helping children read at – Parents valued home
– n = 40 home reading but parents of LD
November 12, 2005
children found it difficult

to help them
– Most common barrier –
16:36
lack of communication
with school
P1: JZZ
Ima and Rumbaut Southeast Asian (SE) 7–12 Btwn group: – CTBS: reading, – SES did not distinguish
0521859751c04a
(1989) LEP & FEP ELLs, – LEP ELLs language between LEP and FEP SE
San Diego school – FEP ELLs – GPA Asian students
district – n = 239 – Students whose parents
– Numeric results but no had more education
stat analyses pre-migration scored
higher on reading than
students whose parents
CB989B/Genesee
had less education at

pre-migration
Jackson and Precocious ELL K, 1 Btwn group: – 9 measures of reading: – The precocious ELL readers
Wen-Hui (1992) readers & – Precocious ELL readers spelling, pseudo-word were similar to
precocious Eng L1 (n = 12) reading, word reading, monolingual precocious
readers, mixed L1 – Precocious English L1 oral reading (speed, readers: all frequently had
0 521 85975 1
backgrounds, readers (n = 12) accuracy); poem someone read to them,

urban & suburban reading, silent reading, discussing the pictures in
settings in the PIAT reading the book, pointing out
Pacific Northwest comprehension (& letters and words, and
others) explaining the meanings of
words; all also had
someone identify numbers
for them, and most had
been helped to identify
letter names, spell words,
and understand word
meanings
– All had attended some kind
of pre-school
November 12, 2005
– All of their parents reported

that the children read at
home at least 2 or 3 times a
16:36
week
173
(continued)
P1: JZZ
0521859751c04a
174
CB989B/Genesee

Kennedy and Park Asian and Mexican 8 Btwn group: – Self-reported English – Mex Am: home language
0 521 85975 1
(1994) American (AM), – Asian Am (n = 1131) course grades use was not related to
diverse – Mexican American (n = – Standardized reading reading, but SES was;
backgrounds 1952) test scores self-control &
– Within ethnic group educational expectations
design with regression were also significantly
analyses related to reading
– Asian Am: home
language use & SES were
related to reading; social
psychological variables
were also significant
Pucci and Ulanoff Hispanic, proficient 4 Btwn group: – Self-report measures of – Proficient readers had
(1998) and less proficient – Proficient readers behavior related to more books in home,
ELL readers, Los (n = 12) reading enjoyed reading more, &
November 12, 2005
Angeles – Less proficient readers (n – Cloze reading test felt they were proficient
= 11) readers than less
– Stats for cloze test proficient readers
16:36
P1: JZZ
0521859751c04a
Reese et al. (2000) Latino, low SES, Los K-7 Within-subject design – Standardized reading – Parental SES was
Angeles – Longitudinal test in Grade 7 (CTBS) significant predictor of
– n = 121 – Parent English reading
– Path analysis interview/survey about – Family literacy practices
family characteristics, & grandparents’ level of
demographics, education also predicted
aspirations & students’ reading
CB989B/Genesee
expectations, role of – ELL’s success in learning

parents in education of to read Eng did not
child, and other factors depend exclusively on L1
– Grade at transition to input
English-only instruction – K. ELLs with greater
emergent Spanish
0 521 85975 1
literacy and Eng oral

proficiency were faster to
transition to Eng reading
& attained higher levels
of English reading
proficiency in middle
school
Thomas and Collier
(2002) See Table 5
for details
Tompkins, ELLs with different 3, 4 Btwn group: – Journal entries: spelling – Students in more affluent
Abramson, and L1s – low SES schools school used more
Pritchard (1999) – English L1 – high SES schools conventional spelling
students, in low- – total n = 60 patterns than students in
November 12, 2005
vs. high-income – statistical & qualitative low SES school

schools, Central analyses
California
16:36
175
P1: JZZ
Academic Achievement
Kathryn Lindholm-Leary and Graciela Borsato
introduction
Academic achievement broadly refers to the communicative (oral, reading,
writing), mathematical, science, social science, and thinking skills and com-
petencies that enable a student to succeed in school and society. Because
these forms of achievement are difficult to assess, most researchers have
relied on a more narrow definition that is largely limited to outcomes
on standardized achievement tests. In this chapter, academic achievement
refers to content-area achievement as measured in English mathematics,
science, or social studies (e.g., history, geography); it does not cover the
content areas of English language arts (addressed in Chapter 4), foreign
language or other humanities (music, art, theater), or cognition (except as
it specifically relates to science or mathematics problem solving). The topic
of reading achievement is included if the outcome measure is a standard-
ized test and the study assesses reading and mathematics achievement of
one or more educational programs. While many of the studies included
in this chapter assess academic achievement by means of standardized
achievement tests, others use general measures of school attainment, such
as grade point average (GPA), high school drop-out rates, and attitudes
toward school and school-related topics.
The academic achievement of ethnic and language minority students has
received considerable attention especially as it relates to the underachieve-
ment of Hispanics, African Americans, Native Americans, and ELLs. This
chapter examines only ELLs and does not consider research on the achieve-
ment of Hispanic, Asian-American, or other ethnic minority or immigrant
students, except as the samples and results pertain to ELLs. As well, studies
that focused only on achievement in Spanish and had no outcome measures
in English were excluded.
This corpus of research is summarized in terms of the following themes
that were identified from our initial review of these studies:
176
P1: JZZ
Academic Achievement 177
r Program Issues
r Language Influences on Academic Achievement
r Instructional Issues
r Family and Learner Background Factors
r Assessment Issues
program issues
Research on the academic achievement of ELLs consists primarily of eval-
uations of various program models. Much of this work addresses policy
issues relating to the best way to educate ELLs. These studies fall into two
distinct groups: (1) evaluation studies (n = 16) that compare student out-
comes on standardized tests of mathematics (and often reading) achieve-
ment across different program types, usually bilingual versus something
else (no program, Structured English Immersion [SEI]/English as a Second
Language [ESL], or two different bilingual models); and (2) evaluation
studies (n = 8) describing student progress in a particular program type,
with outcomes related to standardized achievement tests in mathematics,
science, and social studies, GPA, or high school completion/drop out, or
various school-related attitudes. Most studies (n = 12) concentrate on stu-
dents in elementary school. Five studies focused on high school students
to determine the influence of participation in a bilingual program during
elementary school on their current high school achievement. Studies at the
high school level also often examined GPA, high school drop-out or reten-
tion rates, and attitudes. There were an additional six longitudinal studies
that followed students through elementary school into middle school or
high school and one retrospective study.
Comparative Evaluation Studies

The central issue in the debate on the education of ELLs has been
whether research demonstrates educational benefits of bilingual or other
educational programs specifically designed for ELLs over mainstream
English education. Numerous reviews of the research literature have been
conducted; some reviews have concluded that bilingual programs are inef-
fective and others that bilingual education is more effective than English
immersion approaches (August and Hakuta, 1997). The comparative eval-
uation studies reported here were typically conducted to answer the fol-
lowing two policy-related questions: (1) “Is a language education program
designed for ELLs better than no specially designed program (i.e., main-
stream English)?”, and (2) “Which type of program designed for ELLs leads
to better academic outcomes?” Some studies addressed only one of these
questions and some investigations tackled both. This section summarizes
P1: JZZ
178 Kathryn Lindholm-Leary and Graciela Borsato
the empirical evidence from these studies according to these two questions;
see Table A.5.1 for summaries of the studies.
Question #1: How Effective Are Programs Designed for ELLs in Com-
parison to Mainstream Programs? Beginning in the 1980s, research was
undertaken to address the policy-related question: Is bilingual education as
effective as mainstream education in promoting the academic achievement
and English language proficiency of ELLs? These studies were designed
to compare the academic achievement of ELLs in bilingual education pro-
grams with that of their English-speaking peers in English-only programs.
In most of these studies, comparison students were similar in their ethnic
background (mostly Hispanic; Franco-American in one case) and socio-
economic status (mostly low-income). Results varied according to the
grade level of the students in the sample, the type of program they were in,
and the language and subject area (reading, mathematics) of assessment.
In the aggregate, these studies found that students who had received bilin-
gual instruction performed at grade level usually by the fifth grade, and
they scored at least comparable to if not higher than their English-speaking
peers or their ELL peers who were in mainstream English classes with no
specialized instruction.
As Table A.5.1 shows, there were fifteen studies in this group; five were
conducted in the mid-1980s and ten between 1990 and 2002. In this group
of studies, a wide range of grades was represented, from Grades 1 to 3 to
Grades 7 to 12. In all but six studies (Cazabon, Nicoladis, and Lambert,
1998; de la Garza and Medina, 1985; Gersten and Woodward, 1995;
Lindholm, 1991; Ramirez, 1992; Thomas and Collier, 2002), there was little
or no description of what bilingual instruction entailed, except to indicate
that the curriculum was common across the two programs being com-
pared. Even when the bilingual program was described, the treatment, if
any, that the comparison group received was not always clear (Cazabon,
Lambert, and Hall, 1993; Lindholm, 1991). It was also unclear in many of
these studies how long the students had participated in the bilingual pro-
gram. For example, in Curiel, Rosenthal, and Richek’s (1986) study, stu-
dents in the bilingual group were in the program for at least one year. Thus,
these students could have received one, two, three, or possibly more years
of bilingual education. Yet, several studies clearly show that the amount
of time students participate in a program has a significant impact on their
achievement (Curiel et al., 1986; Gersten and Woodward, 1995).
None of these studies used random assignment; three studies matched
students according to nonverbal intelligence (Cazabon et al., 1993;
Cazabon et al., 1998; Saldate, Mishra, and Medina, 1985), and two for
student or other characteristics, such as same school or district, ethni-
city, social class, program duration (Medrano, 1988), along with birthplace
and education of parents, gender, and family structure (Curiel et al., 1986).
P1: JZZ
In most investigations, comparison groups did not include students who

were individually matched to students in the treatment group. The stu-
dents in most studies were similar in ethnicity, grade level, and social-
class background, except for Burnham-Massey and Piña (1990), who did
not specify the backgrounds of the students included in the comparison
group. Sample sizes were respectable (subjects per group ranged from
forty to five hundred) in all but de la Garza and Medina’s (1985), Fulton-
Scott and Calvin’s (1983), and Saldate et al.’s (1985) (n = 20–31 per group)
research. Thirteen of the fifteen studies used standardized achievement
test scores in English to assess student outcomes (except Curiel et al., 1986,
and Lindholm-Leary and Borsato, 2001).
With respect to reading achievement assessed in English, only one
school site in one study reported that bilingually instructed students per-
formed more poorly than English monolingual comparison students (see
Table A.5.1). Specifically, Alanis (2000) reported that Grade 5 ELLs in
two-way immersion (TWI) programs score significantly lower than EO
peers in reading achievement at one school site; at the other school site
she examined, however, TWI students performed significantly higher in
reading. In four of the remaining studies, bilingually instructed students
scored at levels equivalent to the English-only comparison groups in read-
ing at the final grade level in which comparisons were made (Burnham-
Massey and Piña, 1990; Cazabon et al., 1998; de la Garza and Medina, 1985;
Medrano, 1986). In two studies representing three school sites, students
who were educated through their first language significantly outperformed
monolingually educated comparison students in reading (Saldate et al.,
1985; Thomas and Collier, 2002).
With respect to mathematics achievement, bilingually instructed stu-
dents in four of the studies (Alanis, 2000; Cazabon, Nicoladis, and Lam-
bert, 1998; Medrano, 1986; Saldate et al., 1985) and in one school site in
each of Alanis’ (2000) and Thomas and Collier’s (2002, Site #2 in Table 1)
investigations scored higher than English mainstream students in English
math by the final grade level included in the study. Two studies and one
school site in another study reported that students in the bilingual program
scored at the same level as comparison group students (Burnham-Massey
and Piña, 1990; de la Garza and Medina, 1985; Thomas and Collier, 2002,
Site #1). Bilingually instructed students scored significantly lower than
their English-only peers in only one school site in one study (Alanis, 2000).
Longitudinal and retrospective studies that have assessed high school
students who had been in bilingual programs in elementary school have
also yielded results favoring bilingual programs. Burnham-Massey and
Piña (1990) report an initial lag in performance on the CTBS reading, lan-
guage, and mathematics subtests for students in transitional bilingual edu-
cation (TBE), but found that they caught up to their English-only peers by
Grade 5 and surpassed them in mathematics (but scored comparably in
P1: JZZ
reading) by Grades 7 and 8 (see Table A.5.1). By Grade 11, all students who
have previously been in TBE passed the high school state-level proficiency
subtests. Similar results favoring bilingual programs were obtained in a
study by Curiel et al. (1986), who matched Grades 7 to 10 students on
several background characteristics: gender, birthplace of mother/father,
dual/single-parent family, parental education, and father’s occupation.
They report that ELLs who had participated in a bilingual program in ele-
mentary school had higher GPAs and attendance and a lower high school
drop-out rate than EO Hispanic students who had been in an English main-
stream program during elementary school. The longer the ELLs had had
bilingual instruction, the more positive were their results. Thomas and
Collier (2002) and Lindholm-Leary and Borsato (2001) also report lower
dropout rates for ELLs who had received bilingual instruction in compar-
ison to ELLs in mainstream English classes.
Only one investigation compared the achievement of ELLs in a bilingual
program and ELLs who had had ESL instruction to that of ELLs who
had not participated in a specially designed program (i.e., mainstream
English-only classes) (Thomas and Collier, 2002). In their large-scale study
of four school districts providing specialized instruction to ELLs, Thomas
and Collier (2002) found that, by the end of high school, ELLs who had
been in mainstream classes and had received no special services scored
the lowest in mathematics and reading measured in English and had the
highest dropout rates in high school compared to students who had had
any other type of special services (SEI/ESL, bilingual, two-way).
These studies have been discussed with respect to student performance
in different program types. However, the question remains whether the
students achieved at grade level. As Table A.5.1 indicates, the bilingually
instructed students scored at least at grade level by the final grade in
which they were assessed in English math and reading in five studies
(Burnham-Massey and Piña, 1990; Cazabon et al., 1998; de la Garza and
Medina, 1985; Fulton-Scott and Calvin, 1983; Medrano, 1986). Students
scored average (or close to average) in math but not reading in two stud-
ies (Cazabon et al., 1993; Lindholm, 1991). In one evaluation (Thomas and
Collier, 2002), student achievement varied by school site and program type,
with the bilingually instructed students in the French/English immersion
program performing at grade level. At the second school site, ELLs in
the TBE program scored at grade level; those in the ESL program scored
below grade level; and the mainstream ELLs scored very low. In Gersten
and Woodward’s (1995) study, the TBE and structured English immersion
groups scored below average in language and very low in reading and
vocabulary. In five studies (Alanis, 2000; Curiel et al., 1986; Lindholm-
Leary and Borsato, 2001; Ramirez, 1992; Saldate et al., 1985), either no
achievement data were reported or students’ level of performance was not
clear.
P1: JZZ
In the five investigations in which there was information about Spanish

achievement, students scored at or above grade level in two studies (de
la Garza and Medina, 1985; Thomas and Collier, 2002, Site #2); (slightly)
below grade level in two other studies utilizing the same subjects (Cazabon
et al., 1993; Cazabon et al., 1998); or slightly below average in reading but
slightly above average in math (Lindholm, 1991).
In summary, the question posed in this section can be answered as
follows. Programs that are specially designed for ELLs promote at least
EQUIVALENT and sometimes HIGHER outcomes than a mainstream EO
class. However, if the comparison between a bilingual/ESL/SEI program
and no program/English mainstream is made in the early years of a child’s
education (Grades K-2), it would appear that ELLs in mainstream pro-
grams have an advantage. The extant research also clearly shows that by
the late elementary grades and particularly in middle and high school
grades, ELLs who have received some specialized instruction, particularly
first-language instruction, catch up to and sometimes surpass their com-
parison peers. They score at grade level and are less likely to drop out of
high school. ELLs who had not been in any specialized program but par-
ticipated in mainstream English classes scored the lowest in comparison to
students in any other program and ended their schooling with low levels
of achievement.
Question #2a: Which Model Leads to Higher Academic Outcomes:

Bilingual or ESL or SEI? Ramirez and his colleagues conducted a congres-
sionally mandated four-year longitudinal study of more than two thousand
elementary school ELLs with Spanish language backgrounds (Ramirez,
Yuen, and Ramey, 1991, reported in Ramirez, 1992). The research ques-
tion in this federally sponsored project was: “Which of three alternative
instructional programs designed to meet the needs of Spanish-speaking
LEP students helped them to ‘catch-up’ to their English-speaking peers?”
The three program alternatives were structured English immersion, early-
exit (or TBE), and late-exit bilingual programs. Schools representing the
three models but not individual students were carefully matched so that
the major characteristic that distinguished one program from another was
the amount of L1 instruction provided the students. Students were given
norm-referenced standardized achievement tests, and the Trajectory Anal-
ysis of Matched Percentiles (TAMP) was used to examine the extent to
which students closed the gap with English speakers and made growth
toward grade-level achievement. Ramirez’s study is one of the few studies
that described the instructional model and observed the implementation
of the model under investigation.
Ramirez (1992) reports that the late-exit bilingual program produced
superior outcomes compared to the SEI program, but there were no sig-
nificant differences between the SEI and early-exit (or TBE) bilingual
P1: JZZ
approaches. The study concludes that providing almost exclusive instruc-

tion in English does not accelerate ELLs’ acquisition of English but may
cause them to fall behind; rather, providing ELLs with substantial instruc-
tion in Spanish may assist them in catching up to their English-speaking
peers in English reading, language, and math. Ramirez also reports that
the better the implementation of the model, the stronger were the results
favoring primary-language instruction.
Collier (1992) reanalyzed Ramirez’s data using normal curve equiva-
lents rather than TAMP analysis and confirmed the results reported by
Ramirez. She reports but does not provide data to show that by the end
of Grade 3, students who received no L1 support were beginning to fall
behind, whereas students who were given some L1 support (early-exit)
kept pace with the norm group, though they were not catching up. Only
the late-exit group was able to make greater growth than the norm group
and achieve at grade level. These students were scoring at the 45th NCE
in English reading and the 51st NCE in English math.
Ten years after the Ramirez and Collier reports, Thomas and Collier
(2002) released the results of a large-scale study with mostly native Spanish-
speaking ELLs. Their sample included four research sites representing
rural and urban school districts and different types of language educa-
tion program options for ELLs. Data were collected for students who were
identified as ELL in Grade 1 through the highest grade they had reached
by the time the study terminated. The students had to have participated
in the same program for a minimum of four years. Thus, they assembled
longitudinal data for a large sample of ELLs, including two samples of high
school students. Their results revealed that ELLs enrolled in SEI/ESL pro-
grams scored lower in mathematics and reading achievement measured
in English and had higher school dropout rates compared to students in
bilingual or two-way programs.
Twenty years earlier, Fulton-Scott and Calvin (1983) had reported sim-
ilar findings in a small study of students in a bilingual/bicultural pro-
gram versus a nonintegrated ESL program (ELLs were segregated in ESL
classes) or an integrated ESL program (ELLs received an ESL program
and integrated with their English-speaking peers). Fulton-Scott and Calvin
report that the Grade 6 students who had received bilingual instruction
achieved at a higher level than ELLs who had received ESL, though the
difference was statistically significant only for the comparison between stu-
dents receiving bilingual instruction versus those receiving nonintegrated
ESL instruction.
In the only study to report contrary findings, Gersten and Woodward
(1995) assessed the achievement of ELLs in Grades 4 to 7 of a TBE program
in comparison to that of ELLs in a SEI program (which they called “bilin-
gual immersion”). The sample included only ELLs who had begun their
respective programs in Grade 1 and continued in the program until they
P1: JZZ
were eligible for mainstream instruction. This so-called bilingual immer-

sion (structured English immersion) program included four years dur-
ing which Spanish was used for 30 to 60 minutes per day. Gersten and
Woodward reported that “overall, the data show a consistent pattern. In
the fourth grade, bilingual immersion [SEI] students demonstrated supe-
rior academic performance in all areas assessed. Over time, differences
between the two groups decreased” (p. 232). In other words, they found
differences in Grade 4 right around the time TBE students were transi-
tioning to English, but those differences diminished across Grades 5 to 7,
and by Grade 7 there were no statistically significant differences between
the two groups. A comparison of growth by group revealed that the TBE
students made more progress (growth of 5.2 NCEs) than the SEI students
(growth of 2.4 NCEs), though this difference was not statistically signifi-
cant. Over the four-year time period of the study, results from Tukey post
hoc tests showed significant differences in opposite directions for the two
groups: the total language scores of the SEI students decreased, while the
scores of the TBE students increased. The vocabulary scores remained con-
stant for the SEI students from Grades 4 to 7 while they increased slightly
(3.1 NCEs) from Grades 4 to 7 for the TBE students; this group difference
was not statistically significant.
Question #2b: Which Bilingual Model Leads to Higher Academic

Outcomes: Early Exit or Late Exit? Several studies have compared the
early-exit program model to the late-exit model to determine whether the
greater enrichment and longer duration of L1 instruction in late-exit pro-
grams have a positive, neutral, or negative effect on student academic
achievement. A relevant policy question here is whether one can deliver a
short-term program or whether the program must be longer in duration for
ELLs to demonstrate positive outcomes. Table A.5.1 provides details about
the research methodologies and outcomes of the studies discussed in this
section.
While Ramirez (1992), in his four-year longitudinal study, did not specif-
ically report whether late-exit programs yielded higher outcomes than
early-exit programs, Collier’s (1992) reanalysis of Ramirez’s data did show
higher achievement on the part of students in late-exit programs compared
to early-exit programs. Similarly, Thomas and Collier (2002) report that
late-exit (including two-way) bilingual programs promote higher levels of
reading and mathematics achievement than early-exit (transitional) pro-
grams.
As Table A.5.1 indicates, five studies examined ELLs in two-way pro-
grams in comparison to ELLs in early-exit/TBE programs with respect
to their performance on standardized tests in English reading and math
and their attitudes (Cazabon et al., 1993; Cazabon et al., 1998; Lindholm,
1991; Lindholm-Leary and Borsato, 2001; Thomas and Collier, 2002). At
P1: JZZ
the K-2 grade levels, there was either no difference between early-exit and
two-way immersion programs or a difference favoring early-exit programs
(Cazabon et al., 1993; Lindholm; 1991). However, these same studies con-
sistently reported standardized test results, attitudes, and enrollment in
higher-level math courses favoring two-way programs over early exit/TBE
programs by Grade 3 (Cazabon et al., 1993), Grade 5 (Thomas and Collier,
2002, Site #2), and at high school levels (Cazabon et al., 1998; Lindholm-
Leary and Borsato, 2001).
Ramirez (1992) examined late-exit programs to determine whether stu-
dent outcomes were better for programs that used more Spanish in the
later grades in comparison to programs that used more English in the later
grades. His results showed that students in the school with the greatest
use of Spanish and those in the school that used the most English ended
Grade 6 with comparable levels of skills in English language and read-
ing. However, with respect to mathematics achievement, Grade 6 ELLs
in the two late-exit schools that used more Spanish showed greater gains
in achievement than students at the site that had higher levels of English
instruction at the later grade levels even though both groups of students
had comparable scores in Grade 1. That is, there was a difference in achieve-
ment gains when these models were compared with respect to the amount
of Spanish instruction in the curriculum. ELLs at the late-exit school that
moved abruptly into English instruction (similar to early-exit programs)
showed a marked decline in their growth in mathematics relative to the
norming population. In contrast, ELLs in the late-exit program (particu-
larly at the site that was most faithful to the late-exit instructional model)
showed “continued acceleration in the rate of growth, which is as fast or
faster than the norming population. That is, late-exit students appeared to
be gaining on students in the general population” (p. 25).
In sum, comparisons of ELLs in different program types indicate that
students who received some type of specialized program (TBE, TWI, ESL)
scored higher in reading and math than students who received no special-
ized program in mainstream English classrooms. Also, ELLs who received
some specialized program were able to catch up to and, in some stud-
ies, surpass the achievement levels of their peers who were educated in
English-only mainstream classrooms. This research also indicates that stu-
dents who participated in programs with longer exposure to L1 instruc-
tion (two-way, late-exit) outperformed students who received short-term
exposure (early-exit/TBE). If we consider the interim results for Grades
K-2 (and sometimes 3), we find that bilingually instructed students either
score at lower levels initially than comparison students and perform at
similar levels in higher grades, or they score equivalently in grades K-2
and then outperform comparison-group peers in later grades. This pattern
of results is also evident for other measures of achievement: GPA, school
completion, attitudes toward school, and college preparation. The lower
P1: JZZ
scores in the initial grades may account for the popular misperception that
bilingual education is an ineffective means for educating ELLs.
Descriptive Studies
Several studies have examined and described specific forms of bilingual
programs; these studies are summarized in Table A.5.2. The program model
that has been examined most often is two-way immersion. These studies
have all shown that two-way programs were effective in helping ELLs
achieve at or above grade-level in their L1 and progress toward grade-level
achievement or above in English by middle school. In some of these stud-
ies, the focus was on students’ motivation and attitudes toward learning
(Lambert and Cazabon, 1994; Lindholm-Leary, 2001; Lindholm-Leary and
Borsato, 2001). The results from these studies have shown that ELLs were
developing positive attitudes toward themselves as learners and toward
school. Almost all high school students in two-way programs responded
that they had no intention of dropping out; they wanted to attend a four-
year college and they planned to attend a four-year college right after high
school (Lindholm-Leary and Borsato, 2001).
Seven of these eight studies include results from standardized achieve-
ment tests in reading and math (de Jong, 2002; Kirk-Senesac, 2002; Lambert
and Cazabon, 1994; Lindholm, 1988; Lindholm-Leary, 2001; Lindholm and
Aclan, 1991; Lindholm and Fairchild, 1989) and four included measures of
attitudes (Lambert and Cazabon, 1994; Lindholm, 1988; Lindholm-Leary,
2001; Lindholm-Leary and Borsato, 2001). Three of the studies included
samples of high school students who had or were currently participating in
a two-way program (Kirk-Senesac, 2002; Lindholm-Leary, 2001; Lindholm-
Leary and Borsato, 2001). Most of these studies provide a description of
the two-way model they were studying. Both 90:10 and 50:50 models of
two-way immersion were represented in these studies at both elementary
and high school levels.
These studies consistently show that students score very low in English
reading in the early grades and progress toward grade-level performance
by later elementary or high school. They also show that students who
had developed high levels of proficiency in both languages were more
successful at closing the achievement gap in reading with the norming
group by Grade 4 than students with lower levels of bilingual proficiency
(Kirk-Senesac, 2002; Lambert and Cazabon, 1994; Lindholm-Leary, 2001;
Lindholm and Aclan, 1991). With respect to math achievement, all studies
showed that although the students under evaluation had begun elementary
school with low to below average NCEs, they scored average to above
average in English math by Grades 4 to 6, depending on the study (de
Jong, 2002; Kirk-Senesac, 2002; Lindholm and Aclan, 1991; Lindholm and
Fairchild, 1989). The participating students also typically met district or
P1: JZZ
state proficiency standards (de Jong, 2002; Kirk-Senesac, 2002) and scored
above district and state averages for ELLs (de Jong, 2002; Lindholm-Leary,
2001).
Three other studies examined students’ attitudes (Lambert and
Cazabon, 1994; Lindholm-Leary, 2001; Lindholm-Leary and Borsato, 2001).
These studies found that ELLs had positive perceptions of their academic
competence, bilingualism, and the two-way program itself (Lambert and
Cazabon, 1994; Lindholm-Leary, 2001; Lindholm-Leary and Borsato, 2001).
Lindholm-Leary and Borsato (2001) found that the majority of ELLs were
enrolled in college-bound math courses and only 3 percent in basic math.
Further, most of the students felt that they would not drop out of school;
they wanted to go to college; they would go to college after high school;
and they thought that getting good grades was important. Almost half of
the students felt that the two-way program kept them from dropping out
of school and that they were academically outperforming their peers who
had also started school as ELLs.
The results of these descriptive studies are similar to those presented
in the previous section on comparative studies. That is, students who
received instruction through their L1 closed the achievement gap with
native English speakers and exceeded the performance of their ELL peers
in the district (descriptive studies) or in comparison groups (comparative
studies). In addition, ELLs demonstrated positive attitudes toward the pro-
gram, themselves as learners, school, and other cultures and languages.
Other Program Factors

There has been extensive research in mainstream schools on the character-
istics of effective schools and programs. These studies demonstrate con-
sistently and conclusively that schools with high quality programs have a
cohesive school-wide vision, shared goals that define their expectations for
achievement, a clear instructional focus on and commitment to achieve-
ment, and high expectations (Berman et al., 1995; Corallo and McDonald,
2002; Gándara, 1995; Goldenberg and Sullivan, 1994; Levine and Lezotte,
1995; Marzano, 2003; Montecel and Cortez, 2002; Reyes, Scribner, and
Paredes Scribner, 1999; Slavin and Calderón, 2001; Teddle and Reynolds,
2000; Tikunoff, 1985; U.S. Department of Education, 1998). The importance
of these characteristics has been found in studies of mainstream schools
(e.g., Levine and Lezotte, 1995; Marzano, 2003), low-performing schools
(Corallo and McDonald, 2002; Reyes et al., 1999), and bilingual programs
serving ELLs (e.g., Berman et al., 1995; Montecel and Cortez, 2002; Slavin
and Calderon, 2001; Tikunoff, 1985).
We identified eleven studies that examined program factors associated
with effective schooling for ELLs (Battistich et al., 1997; Berman et al.,
1995; Doherty et al., 2003; Fulton-Scott and Calvin, 1983; Lucas et al., 1990;
P1: JZZ
Mehan et al., 1991; Mehan et al., 1994; Montecel and Cortez, 2002; Ramirez,
1992; Tellez, 1998; Tikunoff, 1985). Two were large-scale federally funded
studies (Berman et al., 1995; Tikunoff, 1985) and one was a small-scale study
that examined the features common to “exemplary” programs that aimed
to promote high achievement among ELLs (Montecel and Cortez, 2002).
Yet other studies looked at the importance of school reform (Goldenberg
and Sullivan, 1994; Minicucci, 1996) or adherence to standards (Doherty
et al., 2003) in promoting academic success among ELLs. Finally, three
studies assessed the effects of integration and “untracking” on high school
students’ achievement (Fulton-Scott and Calvin, 1983; Mehan et al., 1991;
Mehan et al., 1994). Aggregating across this corpus of research, programs
that were relatively effective shared the following characteristics:
r Educational personnel shared the belief that “all children can learn”
(Lucas et al., 1990; Tikunoff, 1985).
r There was a positive school environment; that is, one that was orderly
and safe, had a warm and caring community, and facilitated learning
(Battistich et al., 1997; Berman et al., 1995; Montecel and Cortez, 2002).
r There was a curriculum that was meaningful and academically challeng-
ing, incorporated higher-order thinking (Berman et al., 1995; Doherty
et al., 2003; Montecel and Cortez, 2002; Ramirez, 1992; Tikunoff, 1985),
was thematically integrated (Montecel and Cortez, 2002), and estab-
lished a clear alignment with standards and assessment (Doherty
et al., 2003; Montecel and Cortez, 2002).
r The program model was grounded in sound theory and best practices
associated with an enriched, not remedial, instructional model (e.g.,
Montecel and Cortez, 2002).
r This enriched model was consistent and sustained over time (Ramirez,
1992). Tellez (1998) examined the program placement of close to eleven
thousand ELLs in Grades K-3 and found that one fourth of the students
participated in different program types (as many as three different pro-
gram types in four years). He reported that the greater the assortment of
programs in which an individual student participated, the lower was the
student’s level of achievement. However, he did not provide empirical
data to support this claim.
r Teachers in high quality bilingual programs understood theories about
bilingualism and second-language development as well as the goals and
rationale for the model in which they were teaching (Berman et al., 1995;
Fulton-Scott and Calvin (1983) found that bilingual/bicultural and
integrated ESL programs in which ELLs were integrated with English-
proficient students yielded higher achievement test scores and GPAs than
a segregated ESL program that provided limited opportunities for ELLs to
interact with English-proficient students. These results are consistent with
P1: JZZ
other studies that have shown that achievement levels of ELLs are higher
when they participate in programs that are noncompensatory in nature
and that accord them equal status with other students (Berman et al., 1995;
Mehan et al., 1991; Mehan et al., 1994; Montecel and Cortez, 2002).
These findings on effective programs for ELLs are consistent with
research on effective schools for mainstream students. Both bodies of
research show that students, be they mainstream or ELLs, are more aca-
demically successful when they attend schools that provide positive learn-
ing environments that integrate rather than segregate students and that
include a challenging curriculum for all students. Programs serving ELLs
are more successful when they are based on sound theory; when teach-
ers understand the program model in which they are teaching and the
research-based principles of second-language development that undergird
that model; and when students are placed in a consistent and sustained
program.
language influences on academic achievement

Previous research on bilingualism, which is not part of the present
database, has shown that students with high levels of bilingual profi-
ciency exhibit elevated levels of academic and cognitive functioning in
comparison to students with less well-developed bilingual skills (August
and Hakuta, 1997). A similar conclusion can be drawn from a review of
studies in the present database, which we review in this section. More-
over, it would appear that there is a developmental interconnectedness
among language proficiency, literacy, and content skills within and across
languages for ELLs.
We identified five studies that examined the influence of oral language
proficiency in both L1 and English on math achievement (Fernandez and
Nielsen, 1986; Lindholm-Leary, 2001; Lindholm-Leary and Aclan, 1991;
Nielsen and Lerner, 1986; Rumberger and Larson, 1998). These studies are
consistent in showing that bilingual students with high levels of profi-
ciency in both languages had higher levels of academic achievement than
comparison students who were monolingual. For example, Fernandez and
Nielsen (1986) used the first wave of the High School and Beyond database
(1980), a longitudinal study of U.S. high school seniors and sophomores, to
examine the relationship of bilingualism to academic achievement (educa-
tional expectations, and vocabulary, reading, and math achievement). They
compared the academic performance of Hispanic bilingual (n = 1876), His-
panic English monolingual, Euro-American bilingual, and Euro-American
English monolingual students. Proficiency in both English and the other
language was positively related to achievement for both bilingual groups
(Hispanic bilingual and Euro-American bilingual). More specifically, the
bilingual Hispanics significantly outperformed their English monolingual
P1: JZZ
Hispanic peers in achievement. Nielsen and Lerner also found that frequent
use of the non-English language was negatively associated with achieve-
ment. This finding could be interpreted to mean that the students who
frequently used Spanish lacked sufficient proficiency in English to use that
language more frequently. If this is true, it would reinforce the link between
bilingualism and academic achievement insofar as this latter group of stu-
dents would be classified as nonproficient bilinguals. Nielsen and Lerner
did not define “bilingual” precisely and, indeed, it would appear that it
referred to the students’ exposure to and not necessarily their proficiency
in another language in the home. Nonetheless, this research suggests that
advanced levels of proficiency in two languages may confer advantages
on academic achievement.
Rumberger and Larson (1998) examined the educational achievement
of Grades 7 and 9 Latino students. About 40 percent were identified
as current ELLs, 30 percent as monolingual English speakers (EO), and
30 percent as students who had entered school as ELLs with Spanish as
their home language but were redesignated as FEP speakers. Their results
showed that compared to the EO or ELL students, the FEP students had
higher grades and lower transiency rates, were more likely to be on track
with their high school credits, and showed greater educational commit-
ment. Particularly noteworthy was the finding that the FEP students were
more successful than the monolingual English speakers, despite the fact
that the latter had relatively higher social-class background and might have
been expected to be more proficient in English.
While most of these studies have reported that bilingual students have
higher levels of achievement than monolingual English-speaking students,
only one study in our database examined directly the influence of bilin-
gual proficiency on achievement (Lindholm and Aclan, 1991). More specif-
ically, Lindholm and Aclan (1991) examined a group of ELLs in Grades 1
to 4 in two-way immersion programs in California. They categorized the
students as low, medium, or high bilinguals on the basis of their oral lan-
guage proficiency in each language. The results of academic achievement
testing revealed significant positive relationships between level of bilin-
gual proficiency and level of achievement in math and reading. Moreover,
the students who were classified as “high bilinguals” were able to attain
grade-level results in English reading by Grade 4 and in English math
by Grade 3. Lindholm-Leary (2001) reported similar findings for English
reading, but not math, for Grade 5 ELLs in two-way programs.
Few studies have systematically examined the influence of L1 profi-
ciency or English literacy on content-area achievement. While many of
the comparative evaluation studies reviewed herein discuss reading and
math achievement in each language, most do not examine the relation-
ships between reading achievement in one language with content achieve-
ment in that language or the relationship between content knowledge
P1: JZZ
across languages. Only two U.S.–based studies were identified that address
these relationships; both used correlational techniques. Working in two-
way immersion programs, Lindholm-Leary (2001) found that reading and
math achievement scores were significantly correlated within and across
languages: at Grade 6, English reading with English math (r = 0.81∗∗ ),
Spanish reading with Spanish math (r = 0.74∗∗ ), and Spanish math
with English math (r = 0.85∗∗ ). Moreover, the correlations between math
achievement scores across languages increased across grade levels from
Grade 1 (r = 0.53) to Grade 7 (r = 0.72). Garcia-Vazquez et al. (1997)
also report significant correlations between English reading and math
achievement scores (r = 0.61) in Grades 6 to 12 ELLs. Unfortunately, it
was not clear in what program these students had been or were currently
participating.
instructional issues
In contrast to the plethora of research on instructional issues in literacy (see
Chapter 4), there is a dearth of empirical research on instructional strate-
gies or approaches to teaching content. Only five studies examined instruc-
tional approaches or strategies (Berman et al., 1995; Echevarria, Short, and
Powers, 2003; Minicucci, 1996; Montecel and Cortez, 2002; Tikunoff, 1985),
including one report on schools with exemplary programs in math and
science (Minicucci, 1996). Two other studies, discussed previously, investi-
gated instructional factors that were considered influential in a program’s
effectiveness (Lindholm-Leary, 2001; Ramirez, 1992). We also include here
CREDE technical reports that examined various instructional strategies
that could likely have a significant impact on content learning.
Research on mainstream classrooms shows that good instruction is asso-
ciated with higher outcomes, regardless of the type of educational model
that is used (Levine and Lezotte, 1995; Marzano, 2003; Wenglinsky, 2000).
The same is clearly evident in studies of ELL or high-risk students as well
(Berman et al., 1995; Corallo and McDonald, 2002; Doherty et al., 2003;
Echevarria et al., 2003; Goldenberg and Gallimore, 1991; Montecel and
Cortez, 2002; Ramirez, 1992; Guerrero and Sloan, 2001). Describing good
instruction in programs serving ELLs is complicated because of the con-
stant need to balance the academic and language needs of students with
different levels of language proficiency and, sometimes, with different lan-
guages and cultural backgrounds. Thus, it is even more important to use a
variety of instructional techniques that respond to different learning styles
(Berman et al., 1995; Doherty et al., 2003; Montecel and Cortez, 2002) and
language proficiency levels (Berman et al., 1995; Echevarria et al., 2003;
Montecel and Cortez, 2002). However, which techniques are effective in
producing high-level academic outcomes with ELLs is still an open ques-
tion, as little empirical work has been done on this question.
P1: JZZ
A number of researchers have begun to address this gap in our

knowledge by examining the extent to which incorporating a language-
development component into content instruction results in better access
to the curriculum and higher achievement (Berman et al., 1995; Echevarria
et al., 2003; Minicucci, 1996). Minicucci (1996) found that exemplary schools
for ELLs coordinated math/science instruction with language develop-
ment, thereby providing ELLs with enhanced opportunities to learn and
practice science discourse and writing. In a related vein, Short (1994) has
discussed the importance of explicit language instruction along with con-
tent area instruction. She advocates developing language objectives in
addition to content-area objectives for ELLs to provide them access to the
core curriculum. Echevarria et al. (2003) report that students who were pro-
vided high-quality sheltered instruction, according to the SIOP (Sheltered
Instructional Observation Protocol) model, scored significantly higher and
made greater gains in English writing than comparable ELLs who had not
been exposed to instruction via the SIOP model. While the SIOP model was
developed for ESL teachers to use with ELLs, the approach is clearly appli-
cable to language development with ELLs in various language education
models.
The promotion of positive interactions between teachers and students
is also seen as an important instructional objective. When teachers use
positive social and instructional interactions equitably with both ELL and
EP students, both groups perform better academically (California State
Department of Education, 1982; Doherty et al., 2003). In addition, research
suggests that a reciprocal interaction model of teaching is more beneficial
to students than the traditional teacher-centered transmission model of
teaching (Doherty et al., 2003; Tikunoff, 1985). In the reciprocal interaction
approach, teachers participate in genuine dialogue with students and they
facilitate rather than control student learning. This model encourages the
development of higher-order cognitive skills rather than just factual recall
(Berman et al., 1995; Doherty et al., 2003; Tikunoff, 1985).
High-quality exchanges between teachers and students also give ELLs
better access to the curriculum (Berman et al., 1995; Doherty et al., 2003;
Montecel and Cortez, 2002; Tikunoff, 1985). Minicucci (1996) points out
that exemplary math and science programs in California had teachers who
were proficient in the language of the students or were trained in second-
language acquisition techniques. Stoops Verplaetse highlights this point
in a study on content teachers’ interactions with ELLs (Stoops Verplaetse,
1998). This study showed that teachers issued more directives to and asked
proportionately fewer questions of their ELLs than of their EP students.
Moreover, ELLs were asked fewer high-level cognitive and open-ended
questions. These results confirm findings from studies of teacher talk in
bilingual classrooms by Ramirez (1992) and in two-way classrooms by
Lindholm-Leary (2001) that found that teachers used mostly factual recall
questions and few higher-order cognitive questions with their students.
P1: JZZ
Lindholm-Leary (2001) found that teachers produced the highest percent-

age of higher-order questions when teaching science, though this was not
true of math. When teaching math, as compared to science or social studies,
teachers used a high percentage of directives. Such interaction patterns do
not provide students with opportunities to develop the critical thinking
and discourse skills necessary to succeed in math and science.
In a study of the effectiveness of a collaborative reading approach with
Grade 5 ELLs, Klingner and Vaughn (2000) stress the importance of cooper-
ative and heterogeneous groupings to promote reading acquisition. They
found that students in such collaborative groups spent a considerable
amount of time engaged in academic-related strategic discussion (e.g.,
comprehension checks, elaboration, prompts, feedback). Such grouping
strategies afforded students the opportunity to help each other to under-
stand word meanings related to the content area and to relate their current
knowledge to previous knowledge. Students who participated in these
collaborative groups made statistically significant gains from pre-test to
post-test. Unfortunately, this research did not include a control group of
students who did not receive the treatment. However, this research is con-
sistent with other research showing that cooperative-learning groups pro-
mote higher levels of achievement among ELLs (Calderón and Carreon,
1994; Calderón, Hertz-Lazarowitz, and Slavin, 1998; Calderón, Tinajero,
and Hertz-Lazarowitz, 1992) and that cooperative learning is associated
with effective programs for ELLs (Berman et al., 1995; Doherty et al., 2003;
Research suggests that two other factors promote the educational suc-
cess of ELLs: technology and materials. More specifically, integrating tech-
nology into the curriculum and instruction (Berman et al., 1995; Dixon,
1995) using both languages appears to facilitate student-student inter-
action and, in some cases, learning. Dixon (1995) reports that ELL and
native English-speaking middle-school students could work together effec-
tively on spatial visualization tasks using computers. Moreover, ELLs who
received instruction that integrated technology scored higher on tasks
measuring the concepts of reflection and rotation and on measures of
two-dimensional visualization ability than students who experienced the
traditional textbook approach, and they scored at the same level as the
EP students. With respect to materials, it appears that using a wide vari-
ety of genres of books and many types of materials (visual, audio-visual,
and art materials) improves the success of students in bilingual programs
(Montecel and Cortez, 2002). While instructional materials have received
little empirical attention, August and Hakuta (1997), in reviewing the char-
acteristics of effective schooling for ELLs, found evidence for the impor-
tance of instructional materials appropriate to the needs of these students.
ELLs need access to specialized materials that make the curriculum com-
prehensible to them.
P1: JZZ
family and learner background factors

Several learner and family background factors have been found to influ-
ence student achievement in content-area classes and on standardized tests.
Some studies looked at several different background characteristics and
used regression or correlational analyses to assess the relative effects of
each (Fernandez and Nielsen, 1986; Hampton, Ekboir, and Rochin, 1995;
Nielsen and Lerner, 1986). Other studies selected students with different
background characteristics and looked at group differences on measures of
achievement (Lindholm-Leary, 2001). Many studies examined immigrants;
they did not specifically mention ELLs. The following background charac-
teristics that are related in some way to academic achievement have been
examined: socioeconomic status, length of residence or schooling prior to
entering schools in the United States, and prior knowledge. Each of these
is discussed now.
Socioeconomic Status
An extensive body of research on SES and achievement exists in the main-
stream literature (Knapp and Woolverton, 2003). In contrast, there are rel-
atively few empirical studies of SES and its relationship to achievement
in ELLs. Previously, we noted that most of the literature on ELLs includes
Hispanic students from low-income families. Thus, it is difficult to discern
the true effect of SES because of the limited variation in SES among the
samples that have been examined (Adams et al., 1994). Notwithstanding
this caveat, all studies on ELLs that we reviewed report significant positive
relationships between SES and school outcomes:
r Fernandez and Nielsen (1986) found that SES was an important deter-
minant of achievement in Hispanic bilinguals, but less so than for white
monolinguals.
r Lindholm-Leary (2001) reported significant SES differences in reading
and math achievement for Grade 5 ELLs in two-way immersion pro-
grams with free lunch in comparison to those in non-free lunch pro-
grams.
r Hampton et al. (1995) report that SES was the most significant and dom-
inant predictor of academic performance of 160 Grades 3, 6, and 12 stu-
dents in rural public schools in California.
r Nielsen and Lerner (1986) report that SES and ability were the strongest
determinants of educational expectations among 1,637 Hispanic high
school seniors in the High School and Beyond national survey database.
Another measure of SES that has been used in mainstream and ELL
research is parental level of education (Knapp and Woolverton, 2003).
Adams et al. (1994), using data from the ASPIRA Association Five Cities
P1: JZZ
High School Dropout Study (Velez and Fernandez, 1991), used GPA as the
dependent variable and language proficiency, language dominance, and
parental education as predictor variables. They did not find a significant
correlation between mother’s level of education and student achievement,
arguably because there was little variation in mothers’ levels of education.
Similarly, Lindholm-Leary (2001) found that parental education was not
significantly correlated with the Spanish reading and math achievement
scores of ELLs. However, parental education level correlated significantly
with achievement scores in English reading (NCE = 29 for ≤ high school;
NCE = 50 for college) and math (NCE = 45 for ≤ high school; NCE = 53
for college).
Length of Residence in United States and Amount of Schooling Prior

to Entering U.S. Schools
Research on length of residence in the United States and prior schooling has
yielded some interesting and mixed results. In a study of Hispanic bilin-
gual, Hispanic English monolingual, Euro-American bilingual, and Euro-
American English monolingual students, Fernandez and Nielsen (1986)
found that the longer (more generations) the family had resided in the
United States, the lower the students’ school achievement. Similarly, Adams
et al. (1994) found that more recent immigrants performed at higher levels
than second- or third-generation Hispanic students.
There is an increasing body of literature showing that immigrants
(Hispanic and Asian subgroups) overall, not necessarily ELLs, achieve at
higher levels than second- and later-generation students (Kao and Tienda,
1995; Rumbaut and Portes, 2001). The studies reviewed here concur with
this research. Collier (1987) and Duran and Weffer (1992) add further infor-
mation to this length-of-residence variable.
Duran and Weffer (1992) examined successful Hispanic immigrant high
school seniors (those in the top 25 percent of their class) and found that
number of years in the United States had a positive correlation with read-
ing achievement. The average number of years in the United States was
about nine years for these students, which was measured as the number of
years in the United States when they were in Grade 9. As a group, these stu-
dents performed at grade level, but for each additional year they resided
in the United States, they performed 1.3 NCE points higher in reading
achievement. Length of residence in the United States was unrelated to
math achievement.
In one of the few studies to include ELLs other than Spanish speak-
ers along with a wide range of SES levels, Collier (1987) examined 1,548
ELLs who arrived in the United States at various ages and with a vari-
ety of L1 backgrounds to determine how long it took them to reach
grade-level achievement in ESL classes. All the students had started school
P1: JZZ
as beginning ELLs and remained in the school system for several years.
It is important to point out that students who tested below grade level
in their L1 and older students with little or no formal schooling prior to
entering the United States were excluded from the study. Collier reported
that students who arrived between the ages of eight and eleven were the
fastest achievers; students who arrived at ages five to six were projected to
require at least two to three more years to reach the level of performance
of those who had arrived at ages eight to eleven; and those who arrived
between twelve and fifteen years of age were the lowest achievers and did
not reach the national average in any subject area except math even after
four to five years of residence in the United States. These results are impor-
tant in demonstrating the significance of L1 literacy skills in promoting
achievement in English for students, particularly those who do not arrive
until the later stages of their education (ages twelve to fifteen), where the
academic language demands are very high.
Both Collier (1987) and Duran and Weffer (1992) show the significance
of length of time in the United States. Both, however, only examined stu-
dents who were successful in school. Unfortunately, Duran and Weffer do
not provide enough information about specific length of time in the United
States and whether students who arrived at particular age ranges, as sug-
gested by Collier, were more or less successful than others. However, it
appears from Duran and Weffer’s average age of immigration that their
successful students were not newer arrivals in the age category of twelve
to fifteen, which Collier found to be the least successful. Regardless, these
studies certainly suggest that ELLs who arrive during the high school
years without sufficient English language skills do not appear in the most
successful groups studied and may appear to have the most difficulty in
catching up to grade-level expectations due to the high academic English-
language demands.
Prior Knowledge
While cross-language correlations have been studied extensively in literacy
domains, they have not been examined in domains related to nonlanguage
academic knowledge. However, there is some research on the relationship
between achievement in science and prior knowledge (e.g., playing on a
seesaw and understanding of levers; riding in a boat and understanding
buoyancy). For example, Lee and her colleagues found that prior knowl-
edge and experience (e.g., playing with Legos) provided Hispanic and
Haitian ELLs with more knowledge of science concepts (e.g., concept of
levers) (Lee and Fradd, 1996; Lee, Fradd, and Sutman, 1995). Lee and Fradd
(1996) also found that students with little knowledge of science rarely used
any (or appropriate) strategies to solve science problems and to understand
scientific concepts. In a related vein, Duran, O’Connor, and Smith (1988)
P1: JZZ
found that ELLs’ ability to reason and problem-solve was affected by their
ability to call up relevant cognitive schemata.
Assessment Issues
Assessment plays a central role in the education of all students. Tests are
intended to measure students’ performance in order to identify those who
need assistance and to provide feedback on the effectiveness of instruc-
tional methods and materials. Tests are also used with increasing frequency
to make high-stakes decisions, such as whether a student will move on to
the next grade or receive a diploma, which teachers will receive bonuses,
and whether schools will be rewarded or penalized (Linn, Baker, and
Betebenner, 2002). One of the tenets of the standards-based reform move-
ment is that all children, including ELLs, are expected to attain high stan-
dards. In particular, Title I of the Improving America’s Schools Act (U.S.
Department of Education, 1994) mandates that assessments that determine
the yearly performance of each school must provide for the inclusion of
limited English proficient (i.e., ELL) students. In addition, the No Child
Left Behind Act (U.S. Department of Education, 2001) establishes annual
achievement objectives for ELLs and enforces accountability requirements.
The rationale for including ELLs in high-stakes tests is to hold them to
the same high standards as their peers and to ensure that their needs
are not overlooked (Coltrane, 2002). While there has been an increased
emphasis on the inclusion of ELLs in high-stakes tests, research on assess-
ment that is pertinent to this population is scarce. Only a few studies have
examined this issue that met our inclusion criteria and, as a result, we
have expanded the corpus by including research on testing of ELLs pub-
lished by the national centers associated with educational research (see
Appendix A).
A very important question in the assessment domain is whether English
language proficiency affects performance on academic achievement tests
given in English (August and Hakuta, 1997; Solano-Flores and Trumbull,
2003). If students cannot demonstrate their academic knowledge due to
limited proficiency in English, test results are not valid because they reflect
students’ language skills rather than what they actually know and can do
in academic domains. Results from three studies that have addressed this
issue suggest that, indeed, language affects achievement test performance.
r Alderman (1982) concluded from his study of high school students in
Puerto Rico that there is a risk of underestimating achievement if the test
taker is not proficient in the language of the test. Therefore, proficiency
in the language of testing is a moderator variable that should be taken
into account when assessing academic knowledge.
P1: JZZ
r Pilkington, Piersel, and Ponterotto (1988) found that, for a group of

children with English or Spanish as the home language, the validity of
Kindergarten measures in predicting first-grade achievement varied as
a function of the predominant language in the home.
r Abedi, Lord, and Hofstetter (1998) found that language-related back-
ground variables were good predictors of performance on the National
Assessment of Educational Progress (NAEP) math and reading tests. In
particular, length of time in the United States was the strongest predictor
of students’ performance in math – the longer the students had lived in
the United States, the higher their performance in math.
Stevens, Butler, and Castellon-Wellington (2000) examined the question

of when ELLs are fluent enough to express what they know on a content
test in English. In particular, they studied how the language measured
on the Language Assessment Scales (LAS) Reading Component, a widely
used test of language proficiency, compares to that used on the Iowa Test
of Basic Skills (ITBS) Social Studies Test for Grade 7. They found that the
ITBS contains content-specific academic language, whereas the LAS Read-
ing contains more generic language, common to everyday contexts. They
conclude that the level of language measured by the LAS is not sufficient
to indicate students’ ability to process the language of the ITBS. Moreover,
they found that even though the LAS Reading test has some predictive
validity for the purpose of determining readiness to take content assess-
ments, its ability seems limited to predicting that low scores/high scores on
the LAS will correspond to low/high scores, respectively, on tests such as
the ITBS. In other words, the LAS Reading test lacks fine discriminability.
Several researchers have called into question the appropriateness of
using standardized tests as measures of academic achievement for ELLs by
pointing out that the norming sample for many commonly used tests seri-
ously underrepresents ELLs, rendering norm-referenced scores on these
tests invalid measures of ELLs’ knowledge. Abella (1992) found that Grade
5 students who had exited the English for Speakers of Other Languages
(ESOL) Program were, according to their grades, performing successfully
in school, even though they scored well below the national average on most
subtests of the Stanford Achievement Test. In contrast, Davidson (1994)
found that the use of The 3-R’s (a nationally standardized test normed on
English speakers) with ELLs did not represent a serious statistical viola-
tion of test norms. At the same time, he went on to argue that multiple
measures should be used in order to compensate for the misapplication of
norm-referenced tests. Butler et al. (2000) have pointed out that the Stan-
ford 9 (SAT-9) test is not an appropriate measure to assess the academic
achievement of ELLs because it was designed to distinguish levels of aca-
demic achievement among native English speakers.
P1: JZZ
Linn et al. (2002) note that the process of establishing proficiency stan-
dards for state tests has lacked rigor. Review panels, that typically included
teachers and often other interested citizens, reviewed tests and identified
cutoff scores that they thought would “correspond to the level of perfor-
mance expected from a proficient student who is motivated to do well and
has had an adequate opportunity to learn the material” (p. 4). They go on
to note that the outcome of such a process has led to the establishment
of proficiency levels that are so high they are unrealistic. Moreover, when
these proficiency standards were developed, the educators who developed
them were unaware that the standards would be used to determine Ade-
quate Yearly Progress (AYP) objectives or that sanctions would be imposed
if they did not satisfy AYP. From an accountability perspective, a problem
arises because the definition of “passing” varies tremendously from state to
state. Linn et al. (2002) suggest using “scale scores and monitoring changes
in average scores over time in terms of standard deviation units, thereby
avoiding the need for performance standards altogether” (p. 16).
Royer and Carlo (1991) have argued that traditional measures of aca-
demic achievement should not be used with ELLs because of mismatches
between test content on the one hand and educational, cultural, and
linguistic experience on the other. They advocate the use of alternative
methods to assess ELLs and, in particular, they propose the use of the
Sentence Verification Technique (SVT). The SVT consists of reading or lis-
tening to a passage and then responding “Yes” or “No” to test sentences
depending on whether they correctly reflect the information in the pas-
sage. Royer and Carlo found that the reliability and validity of SVT tests
are very good. Other researchers, including Valdez-Pierce and O’Malley
(1992), have recommended the use of assessment procedures, such as oral
interviews, teacher observation checklists, student self-evaluations, and
portfolio assessment, that reflect tasks typical of the classroom; these are
referred to as authentic assessment. Because authentic assessment strategies
ask students to show what they can do, it is argued they provide a more
accurate measure of what students know, independent of their language
skills, than standardized tests.
An important line of research regarding the use of standardized assess-
ments with ELLs concerns test accommodations. Accommodations for
ELLs have been defined as “the support provided students for a given test
event . . . to help [them] access the content in English and better demon-
strate what they know” (Butler and Stevens, 1997, p. 5). Abedi (2001) has
pointed out that “accommodations are intended to level the playing field,
that is, to make language less of a factor, or ideally a non-factor, when mea-
suring performance” (p. 2). Despite the potential importance of test accom-
modations, empirical studies informing the use of them with ELLs are
scarce. Findings from these studies, which are summarized herein, suggest
that the language of assessment should match the language of instruction
P1: JZZ
and that modifying test questions to reduce language complexity may

help narrow the performance gap between native English speakers and
ELLs.
Abedi et al. (1998) randomly assigned Grade 8 ELLs to one of three ver-
sions of the NAEP math assessment: original English language, modified
English language (the nontechnical language in the original test was made
more readily understandable), and original Spanish language. They found
that students performed highest on the modified English version, lower on
the original English version, and lowest on the original Spanish version.
These results suggest that clarifying the language of the test and testing in
the language of instruction helps students improve their test performance.
Abedi et al. (2001a) examined the impact of accommodations on the per-
formance of Grade 8 students taking the NAEP mathematics assessment.
Five different forms of the test were randomly distributed: original English
language (items taken directly from the NAEP test booklet), linguistically
simplified English, original English language accompanied with a glossary
for the potentially difficult terms, original English language version with
extra time to complete the test, and original English language version with
a glossary and extra time. Their results indicate that most accommodations
helped both ELL and non-ELL students, with the greatest score improve-
ments resulting from the version that included a glossary and allowed
extra time. The only accommodation that narrowed the performance gap
between ELL and non-ELL students was the simplified English version of
the test.
Abedi et al. (2001b) conducted a pilot study of three forms of twenty
NAEP items from a Grade 8 science test: original items and no accommoda-
tion, original items and customized English language dictionary at the end
of the test booklet, and original items accompanied by English glosses and
Spanish translation of the glosses in the margins of the test booklet. They
found that ELLs performed substantially higher under the accommodated
conditions than under the standard conditions, and they did particularly
well under the customized dictionary condition. Moreover, the accommo-
dations did not have a significant effect on the scores of the non-ELLs,
suggesting that the accommodations did not alter the construct under
measurement.
Abedi and Lord (2001) compared the performance of ELLs and pro-
ficient speakers of English on math problems from the NAEP tests and
examined whether modifying the original items to reduce language com-
plexity would impact student performance. Overall, the ELLs obtained
lower scores than the proficient speakers of English. Scores on the linguis-
tically modified version of the test were slightly higher. The language mod-
ifications had especially beneficial effects on the results of low-performing
students: ELLs benefited more than proficient speakers of English; low SES
students benefited more than others; and students in low-level and average
P1: JZZ
math classes benefited more than those in high-level math and algebra
classes.
Shepard, Taylor, and Betebenner (1998) found that accommodations con-
sistently raised the relative position of ELLs on the Rhode Island Grade
4 Mathematics Performance Assessment. Accommodations consisted, for
the most part, of a change in the conditions of test administration, such as
oral reading of the assessment or extended time to complete it.
Castellon-Wellington (2000) investigated the effects of two types of
test accommodations: providing extra assessment time and reading items
aloud, on the standardized test performance of Grade 7 ELLs. Her results
indicated that students did not significantly improve their performance on
the Iowa Tests of Basic Skills (ITBS) Social Studies Test as a result of either
accommodation, even when given their preferred accommodation.
In summary, both federal and state legislation requires inclusion of all
students, including ELLs, in large-scale content assessments. Inclusion of
ELLs “signals the commitment of the educational system to support the
academic progress of all its students; and it ensures the representativeness
of the data reported” (Shepard et al., 1998, p. 1). The use of the same stan-
dardized assessments with all students is, however, problematic. Unless
the students on whom the test was standardized match closely the students
who take the test, the test norms may be inappropriate for the underrepre-
sented students and the resulting scores could be unreliable and possibly
invalid (Davidson, 1994). Moreover, the research reviewed here indicates
that language proficiency and other background factors (e.g., length of
stay in the United States) strongly relate to test performance, indicating
additional sources of invalidity. Results from the few studies that have
investigated accommodations suggest that the language of assessment
should match the language of instruction and that modifying test ques-
tions to reduce language complexity may help narrow the performance
gap between native English speakers and ELLs.
conclusions
Before summarizing the major findings from these studies, it is important
to address the question of the generalizability of this body of research.
A major methodological limitation concerns the kinds of students who
have been evaluated in this body of research. Except for three studies that
included Asian language background students and one study with Franco-
American students, all study participants were Hispanic low-income stu-
dents. That most of these students are from low-income backgrounds is
important to stress because these findings might not be applicable to His-
panics from families with higher incomes and parental education. A corol-
lary problem concerns the participants in the comparison group who, for
many of the studies, were appropriately matched in terms of ethnicity and
P1: JZZ
SES with those in the intervention group. However, while this is method-
ologically sound, it is important to keep in mind that the outcomes for these
comparison groups may not reflect the average performance of English-
only students in mainstream classrooms.
Another major problem is that the definitions of various program mod-
els are often vague. In some cases, bilingual education is clearly defined
with respect to amount of instruction time devoted to each language and
length of duration of the program (e.g., early-exit or transitional; late-exit
or maintenance) (e.g., Ramirez, 1992; Thomas and Collier, 2002). In other
cases, it is not clear what specialized instruction the students received
in their “bilingual” classroom (Burnham-Massey and Piña, 1990; Curiel
et al., 1986; Medrano, 1988; Saldate et al., 1985). In nonbilingual con-
texts, sometimes a mainstream English classroom was labeled “structured
English immersion” and, in other contexts, structured English immer-
sion included specialized instruction for ELLs (Ramirez, 1992). Further,
Gersten and Woodward (1995) used the term “bilingual” immersion to
refer to English immersion that included daily instruction through Span-
ish but was interpreted as English-only instruction when compared to
the alternative transitional bilingual education program. Yet, Gersten and
Woodward’s bilingual immersion was similar to what many have called
structured English immersion or just English immersion, while the term
bilingual immersion has been used by other authors to refer to two-way
(bilingual) immersion education (Lindholm-Leary, 2001; Thomas and Col-
lier, 2002). These methodological and definitional differences have added
fuel to the debate on the effectiveness of alternative educational programs
for ELLs.
Turning to general findings – with respect to program issues, there is
strong convergent evidence that the educational success of ELLs is posi-
tively related to sustained instruction through the student’s first language.
In both the descriptive and comparative program evaluation studies,
results showed that length of time in the program and time of assessment
affect outcomes. Evaluations conducted in the early years of a program
(Grades K-3) typically reveal that students in bilingual education scored
below grade level (and sometimes very low) and performed either lower
than or equivalent to their comparison group peers (i.e., ELL students in
mainstream English, SEI/ESL, or EO students in mainstream classrooms).
Almost all evaluations of students at the end of elementary school and in
middle and high school show that the educational outcomes of bilingually
educated students, especially in late-exit and two-way programs, were at
least comparable to and usually higher than their comparison peers. There
was no study of middle school or high school students that found that
bilingually educated students were less successful than comparison-group
students. In addition, most long-term studies report that the longer the stu-
dents stayed in the program, the more positive were the outcomes. These
P1: JZZ
results hold true whether one examines outcomes in reading or mathe-

matics achievement, GPA, attendance, high school completion, or attitudes
toward school and self. Further, in one study (Lindholm-Leary and Borsato,
2001), half of the ELLs attributed their positive attitudes toward staying in
school to the two-way program in which they participated.
These findings about the importance of early first-language instruc-
tion are also consistent with Collier’s (1987) research on immigrant stu-
dents who arrived in the United States at different ages. More specifically,
her findings showed that those students who immigrated after they had
received formal schooling in their home country, having already developed
strong first-language and academic skills, were more likely to close the gap
with native English-speaking U.S. students in English achievement than
students who had immigrated at younger ages and were placed in a main-
stream English classroom before they had developed a strong conceptual
base in any language. One additional important point was exemplified in
the research by Tellez (1998), who reported that students who participated
in a hodgepodge of different programs performed at the lowest educa-
tional levels. In the longitudinal studies discussed here (e.g., Thomas and
Collier, 2002), students who received no intervention performed at the
lowest levels and had the highest drop-out rates.
Overall, research is consistent in showing that ELLs who received some
specialized program (TBE, TWI, or ESL) were able to catch up to and
in some studies surpass the achievement levels of their ELL peers and
their English-only peers who were educated in English-only mainstream
classrooms. These findings indicate further that ELLs who participated
in programs that provided extended instruction through the medium of
the students’ L1 (i.e., two-way immersion and late-exit program) outper-
formed students who received short-term instruction through their L1 (i.e.,
early-exit and TBE programs).
Nothwithstanding these positive results, they must be tempered in light
of methodological concerns. As noted earlier, only a couple of studies used
random assignment of students to treatment and control groups. As Willig
(1985) noted, random assignment usually resulted in stronger effect sizes
for bilingual over monolingual instruction. Thus, the results summarized
here may actually underestimate the true advantages of bilingual instruc-
tion. At the same time, and as pointed out in Chapter 4, there are problems
associated with random assignment. In addition to ethical issues and gener-
alizability problems associated with random assignment, there is the issue
of random assignment of teachers to classrooms. More specifically, unless
teachers are randomly assigned as well, a clearly unrealistic possibility,
random assignment of students will not provide the methodological clar-
ity that is sought by critics of this body of research. Another shortcoming
of this research concerns the lack of precise definitions of program models
and inconsistent use of terminology. Programs were often simply labeled
P1: JZZ
“bilingual” or “ESL” without describing what was actually going on in the

program. For example, there was no or poor specification of the amount of
instruction in each language or the amount of specialized English instruc-
tion, the duration of treatment, and the language proficiency of teachers.
There is clearly a need for carefully controlled studies that consider what
program models work under what conditions and for whom.
The studies reviewed here also indicate that bilingual proficiency and
biliteracy are positively related to academic achievement in both lan-
guages. More specifically, bilingual Hispanic students had higher achieve-
ment scores (Fernandez and Nielsen, 1986; Lindholm-Leary, 2001; Nielsen
and Lerner, 1986; Rumberger and Larson, 1998), GPAs, and educational
expectations (Fernandez and Nielsen, 1986; Nielsen and Lerner, 1986) than
their monolingual English-speaking Hispanic peers. In addition, there
were significant positive correlations between Spanish reading and English
reading, between English reading and English math, and between Spanish
reading and Spanish math, suggesting that there are complex but support-
ive interdependencies in the language, literacy, and academic development
of bilingual students. These results suggest that educational programs for
ELLs should seek to develop their full bilingual and biliterate competen-
cies in order to take advantage of these developmental interdependen-
cies. At the same time, it is important to point out that while the research
findings reported here are consistent with one another and with previous
reviews (August and Hakuta, 1997; see Chapter 3), the actual research base
is scant and consists mostly of correlational studies. A systematic program
of empirical research is needed to determine how the two languages inter-
act and how and under what conditions content instruction through one
language facilitates content acquisition in another language.
Research reviewed here has identified a number of instructional char-
acteristics that are influential in promoting the academic success of ELLs.
Many of these characteristics have also been identified in the success of
mainstream students, including cooperative learning, integrating technol-
ogy into the curriculum, high-quality exchanges between teachers and
pupils, and responsiveness to different learner styles. What distinguishes
research on ELLs from that on mainstream students is the importance
of making the curriculum accessible to students who are not fully pro-
ficient in English. Indeed, research indicates the importance of incorpo-
rating language-development components and sheltering techniques into
content instruction.
While research has examined the characteristics of effective content
instruction, most of these findings emanate from studies in which the
primary focus was not on the instructional characteristics. There is little
research into how to make instruction more accessible and meaningful to
ELLs in areas considered challenging by native English speakers – that
is, science and math. Extant research provides some starting points, but
P1: JZZ
a research program that includes students with different language back-

grounds, ages, and previous educational experiences prior to immigrating
should be considered in future research. For example, what approaches
can assist new immigrant students at the late elementary or high school
level to learn content when they have never experienced school in their
home country? Collier’s (1987) research showing the importance of previ-
ous instruction in the country of origin on students’ academic success in
the United States demonstrates the importance of including this variable.
A significant factor to consider in examining such learners is the cognitive
overload they experience when learning academic content area through a
second language. With the increasing significance of technology in soci-
ety, research should also include systematic investigation of the impact of
computer-assisted instruction for ELLs. Dixon’s (1995) research is instruc-
tive in demonstrating that technology provides a significant vehicle for
successful learning, especially with helpful peers. Finally, research on the
influence of peer interactions in promoting content learning in ELLs would
provide information on additional contexts in which to help ELLs develop
language and content skills, a point that was raised in Chapter 2.
In addition, socioeconomic status, length of residence in the United
States, and prior knowledge can be influential factors in ELLs’ academic
achievement and success in school. Similar to research on mainstream
students, research with ELLs found that there is a positive relation-
ship between a student’s level of SES and their level of achievement;
recent immigrants are more academically successful than second- or third-
generation students; and prior knowledge influences academic achieve-
ment. The paucity of studies in this area is a significant gap in the empiri-
cal literature on learner and family background factors that may inhibit or
promote the academic development of ELLs.
As the reviewed research also showed, there are many challenges in
assessing the content-area knowledge of ELLs. Test norms may be inap-
propriate because of differences between ELLs and students in the norm-
ing samples; and language proficiency and other background factors may
influence test performance. Results from the few studies that have inves-
tigated testing accommodations suggest that the language of assessment
should match the language of instruction and that modifying test ques-
tions to reduce language complexity may help narrow the performance gap
between native English speakers and ELLs. Given the high-stakes nature
of standardized testing, it is imperative to develop assessment procedures
that allow all students to demonstrate what they know.
Taken together, these results indicate that ELLs are more successful
when they participate in programs that are specially designed to meet their
needs (ESL/SEI, bilingual) than in mainstream English classrooms and
when the program is consistent throughout the student’s education. A pro-
gram that is enriched, consistent, and provides a challenging curriculum
P1: JZZ
is also endorsed by research on factors associated with effective programs

for ELLs.
Much current research suffers from a short-term perspective; only a
few studies have examined the long-term results of various program
alternatives. Retrospective studies and those that rely wholly on school
records, while valuable, provide incomplete information, and certainly
cannot address the complex teacher and instructional factors that can affect
student outcomes. Indeed, extant research fails to include sufficient infor-
mation on the specific teacher and instructional factors that are associated
with and presumably responsible for successful content learning. More-
over, the oft-asked question, “Which program is best for ELLs?” is overly
simplistic because it assumes that only one approach is the best for all
students under all circumstances.
References
Abedi, J. 2001. Assessment and accommodations for English language learners: Issues
and recommendations (Policy Brief 4). Los Angeles: University of California,
National Center for Research on Evaluation, Standards, and Student Testing.
Abedi, J., Hofstetter, C., Baker, E., and Lord, C. 2001a. NAEP Math performance
and test accommodations: Interactions with student language background (CSE
Technical Report 536). Los Angeles: University of California, National Cen-
ter for Research on Evaluation, Standards, and Student Testing.
Abedi, J., and Lord, C. 2001. The language factor in mathematics tests. Applied
Abedi, J., Lord, C., Boscardin, C. K., Miyoshi, J. 2001b. The effects of accommodations
on the assessment of limited English proficient (LEP) students in the National Assess-
ment of Educational Progress (NAEP) (CSE Technical Report 537). Los Angeles:
University of California, National Center for Research on Evaluation, Stan-
dards, and Student Testing.
Abedi, J., Lord, C., and Hofstetter, C. 1998. Impact of selected background variables on
students’ NAEP math performance (CSE Technical Report 478). Los Angeles: Uni-
versity of California, National Center for Research on Evaluation, Standards,
and Student Testing.
Abella, R. 1992. Achievement tests and elementary ESOL exit criteria: An evalua-
tion. Educational Evaluation and Policy Analysis 14(2), 169–74.
Adams, D., Astone, B., Nunez-Wormack, E. M., and Smodlaka, I. 1994. Predicting
the academic achievement of Puerto Rican and Mexican-American ninth-grade
students. Urban Review 26(1), 1–14.
Alanis, I. 2000. A Texas two-way bilingual program: Its effects on linguistic and
academic achievement. Bilingual Research Journal 24(3), 225–48.
Alderman, D. L. 1982. Language proficiency as a moderator variable in testing
academic aptitude. Journal of Educational Psychology 74(4), 580–7.
Aspiazu, G. G., Bauer, S. C., and Spillett, M. 1998. Improving the academic perfor-
mance of Hispanic youth: A community education model. Bilingual Research
Journal 22(2–4), 127–47.
P1: JZZ
August, D., and Hakuta, K. (eds.). 1997. Improving schooling for language minority
children: A research agenda. Washington, DC: National Academy Press.
Baker, K. A., and de Kanter, A. A. 1981. Effectiveness of bilingual education: A review
of the literature. Washington, DC: U.S. Department of Education, Office of Plan-
ning, Budget and Evaluation.
Battistich, V., Solomon, D., Watson, M., and Schaps, E. 1997. Caring school commu-
nities. Educational Psychology 32, 137–51.
Berman, P., Minicucci, C., McLaughlin, B., Nelson, B., and Woodworth, K. 1995.
School reform and student diversity: Case studies of exemplary practices for English
language learner students. Santa Cruz, CA: National Center for Research on
Cultural Diversity and Second Language Learning, and B. W. Associates.
Burnham-Massey, L., and Piña, M. 1990. Effects of bilingual instruction on English
academic achievement of LEP students. Reading Improvement 27(2), 129–32.
Butler, F. A., and Stevens, R. 1997. Accommodation strategies for English language
learners on large-scale assessments: Student characteristics and other considerations
(CSE Technical Report 448). Los Angeles: University of California, National
Center for Research on Evaluation, Standards, and Student Testing.
Butler, Y. G., Orr, J. E., Bousquet-Gutiérrez, M., and Hakuta, K. 2000. Inadequate
conclusions from an inadequate assessment: What can SAT-9 scores tell us
about the impact of Proposition 227 in California? Bilingual Research Journal
24(1–2), 141–54.
Calderón, M., and Carreon, A. 1994. Educators and students use cooperative learn-
ing to become biliterate and bicultural. Cooperative Learning Magazine, 4, 6–9.
Calderón, M., Hertz-Lazarowitz, R., and Slavin, R. 1998. Effects of bilingual coop-
erative integrated reading and composition on students making the transition
from Spanish to English. The Elementary School Journal, 99, 153–65.
Calderón, M., Tinajero, J., and Hertz-Lazarowitz, R. 1992. Adapting cooperative
integrated reading and composition to meet the needs of bilingual students.
Journal of Educational Issues of Language Minority Students 10, 79–106.
California State Department of Education. 1982. Basic Principles for the Education
of Language Minority Students, An Overview. Sacramento: Office of Bilingual
Bicultural Education.
Castellon-Wellington, M. 2000. The impact of preference for accommodations: The per-
formance of English language learners on large-scale academic achievement tests (CSE
Technical Report 534). Los Angeles: University of California, National Center
for Research on Evaluation, Standards, and Student Testing.
Cazabon, M., Lambert, W., and Hall, G. 1993. Two-way bilingual education: A progress
report on the Amigos Program (Research Report No. 7). Santa Cruz, CA: National
Center for Research on Cultural Diversity and Second Language Learning.
Cazabon, M., Nicoladis, E., and Lambert, W. E. 1998. Becoming bilingual in the Amigos
two-way immersion program (Research Report No. 3). Santa Cruz, CA: Center
Collier, V. P. 1987. Age and rate of acquisition of second language for academic
1992. A synthesis of studies examining long-term language minority student
data on academic achievement. Bilingual Research Journal 16(1–2), 187–212.
Coltrane, B. 2002. English-language learners and high-stakes tests: An overview of
the issues. Washington, DC: Center for Applied Linguistics (ERIC Document
Reproduction Service No. EDO-FL-02–07).
P1: JZZ
Corallo, C., and McDonald, D. H. 2002. What works with low-performing schools:
A review of research. Charleston, WV: AEL, Regional Educational Laboratory,
Region IV Comprehensive Center.
Curiel, H., Rosenthal, J. A., and Richek, H. G. 1986. Impacts of bilingual education
on secondary school grades, attendance, retentions and drop-out. Hispanic
Journal of Behavioral Sciences 8(4), 357–67.
Davidson, F. 1994. Norms appropriacy of achievement tests: Spanish-speaking chil-
dren and English children’s norms. Language Testing 11(1), 83–95.
de Jong, E. J. 2002. Effective bilingual education: From theory to academic achieve-
ment in a two-way bilingual program. Bilingual Research Journal 26(1), 65–84.
de la Garza, J. V., and Medina, M. 1985. Academic achievement as influenced by
bilingual instruction for Spanish-dominant Mexican American children. His-
panic Journal of Behavioral Sciences 7(3), 247–59.
Dixon, J. K. 1995. Limited English proficiency and spatial visualization in mid-
dle school students’ construction of the concepts of reflection and rotation.
Bilingual Research Journal 19(2), 221–47.
Doherty, R. W., Hilberg, R. S., Pinal, A., and Tharp, R. G. 2003. Five standards and
student achievement. NABE Journal of Research and Practice 1(1), 1–24.
Dulay, H., and Burt, M. 1980. The relative proficiency of limited English proficient
students. In J. E. Alatis (ed.), Current Issues in Bilingual Education (pp. 181–200).
Washington, DC: Georgetown University Press.
Duran, B. J., and Weffer, R. E. 1992. Immigrants’ aspirations, high school process,
and academic outcomes. American Educational Research Journal 29(1), 163–81.
Duran, R. P., O’Connor, C., and Smith, M. 1988. Methods for assessing reading com-
prehension skills of language minority students (Project 2.2 Technical Report).
Los Angeles: University of California, Center for Language Education and
Research.
Duran, R. P., Revlin, R., and Havill, D. 1995. Verbal comprehension and reasoning skills
of Latino high school students (Research Report No. 13). Santa Cruz, CA: National
Center for Research on Cultural Diversity and Second Language Learning.
Echevarria, J., Short, D., and Powers, K. 2003. School reform and standards-based
education: How do teachers help English language learners? Technical report.
Santa Cruz, CA: Center for Research on Education, Diversity & Excellence.
Fernandez, R., and Nielsen, F. 1986. Bilingualism and Hispanic scholastic
achievement: Some baseline results. Social Science Research 15, 43–70.
Fulton-Scott, M. J., and Calvin, A. D. 1983. Bilingual multicultural education
vs. integrated and non-integrated ESL instruction. NABE: The Journal for the
National Association for Bilingual Education 7(3), 1–12.
Gándara, P. 1995. Over the ivy walls: The educational mobility of low-income Chicanos.
New York: State University of New York Press.
Gándara, P., and Merino, B. 1993. Measuring the outcomes of LEP programs: Test
scores, exit rates, and other mythological data. Educational Evaluation and Policy
Analysis 15(3), 320–38.
Garcı́a-Vázquez, E., Vázquez, L. A., López, I. C., and Ward, W. 1997. Language
Garcı́a-Vázquez, E., Vázquez, L. A., López, I. C., & Ward, W. 1999. Language
P1: JZZ

Gersten, R., and Woodward, J. 1995. A longitudinal study of transitional and immer-
sion bilingual education programs in one district. Elementary School Journal
95(3), 223–39.
Goldenberg, C. N., and Gallimore, R. 1991. Local knowledge, research knowledge,
and educational change: A case study of early Spanish reading improvement.
Educational Researcher 20, 2–14.
Goldenberg, C. N., and Sullivan, J. 1994. Making change happen in a language-
minority school: A search for coherence, (Educational Practice Report #13).
Santa Cruz: University of California at Santa Cruz, National Center for
Research on Cultural Diversity and Second Language Acquisition.
Guerrero, M., and Sloan, K. 2001. A descriptive analysis of four exemplary K-
3 Spanish reading programs in Texas: Are they really exemplary? Bilingual
Research Journal 25, 253–80.
Hampton, S., Ekboir, J. M., Rochin, R. I. 1995. The performance of Latinos in rural
public schools: A comparative analysis of test scores in grades 3, 6, and 12.
Hispanic Journal of Behavioral Sciences 17(4), 480–98.
Kao, G., and Tienda, M. 1995. Optimism and achievement: The educational perfor-
mance of immigrant youth. Social Science Quarterly 76, 1–19.
Kirk-Senesac, B. V. 2002. Two-way bilingual immersion: A portrait of quality school-
ing. Bilingual Research Journal 26(1), 85–101.
Klingner, J., and Vaughn, S. 2000. The helping behaviors of fifth graders while using
collaborative strategic reading during ESL content classes. TESOL Quarterly
34(1), 69–98.
Knapp, M. S., and Woolverton, S. 2003. Social class and schooling. In J. Banks and
C. A. McGee Banks (eds.), Handbook of research on multicultural education, 2nd
Edition (pp. 548–69). New York: Jossey Bass.
Lambert, W. E., and Cazabon, M. 1994. Students’ view of the Amigos program (Research
Report No. 11). Santa Cruz, CA: National Center for Research on Cultural
Lee, O., and Fradd, S. H. 1996. Literacy skills in science learning among linguisti-
cally diverse students. Science Education 80(6), 651–71.
Lee, O., Fradd, S. H., and Sutman, F. X. 1995. Science knowledge and cognitive
strategy use among culturally and linguistically diverse students. Journal of
Research in Science Teaching 32(8), 797–816.
Levine, D. U., and Lezotte, L. W. 1995. Effective schools research. In J. A. Banks
and C. A. Mcgee Banks (eds.), Handbook of research on multicultural education
(pp. 525–47). New York: Macmillan.
Lindholm, K. J. 1988. The Edison elementary school bilingual immersion program: Student
progress after one year of implementation. Technical Report No. 9 of the Center for
Language Education and Research, UCLA.
1991. Theoretical assumptions and empirical evidence for academic achieve-
ment in two languages. Hispanic Journal of Behavioral Sciences 13(1), 3–
17.
Lindholm, K. J., and Aclan, Z. 1991. Bilingual proficiency as a bridge to academic
173(2), 99–113.
P1: JZZ
Lindholm, K., and Fairchild, H. 1989. Evaluation of an “exemplary” bilingual immersion

program (Technical Report TR13). Los Angeles, CA: University of California,
Center for Language Education and Research (ERIC Document Reproduction
Service No. ED 307 820).
Lindholm-Leary, K. J. 2001. Dual language education. Avon, UK: Multilingual
Matters.
Lindholm-Leary, K. J., and Borsato, G. 2001. Impact of two-way bilingual elementary
programs on students’ attitudes toward school and college. Santa Cruz, CA: Center
Linn, R. L., Baker, E. L., and Betebenner, D. W. 2002. Accountability systems: Impli-
cations of requirements of the No Child Left Behind Act of 2001. Educational
Researcher 31(2), 3–16.
Lucas, T., Henze, R., and Donato, R. 1990. Promoting the success of Latino language-
minority students: An exploratory study of six high schools. Harvard Educa-
tional Review 60(3), 315–40.
Marzano, R. J. 2003. What works in schools: Translating research into action. Alexandria,
VA: Association for Supervision and Curriculum Development.
Medrano, M. F. 1986. Evaluating the long-term effects of a bilingual education
program: A study of Mexican-American students. Journal of Educational Equity
and Leadership 6(2), 129–38.
1988. The effects of bilingual education on reading and mathematics
achievement: A longitudinal case study. Equity and Excellence 23(4), 17–19.
Mehan, H., Datnow, A., Bratton, E., Tellez, C., Friedlaender, D., and Ngo, T. 1991.
Untracking and college enrollment (Research Report No. 4). Santa Cruz, CA:
Center for Research on Education, Diversity & Excellence.
Mehan, H., Hubbard, L., Lintz, A., and Villanueva, I. 1994. Tracking untracking: The
consequences of placing low track students in high track classes (Research Report
No. 10). Santa Cruz, CA: National Center for Research on Cultural Diversity
and Second Language Learning.
Minicucci, C. 1996. Learning science and English: How school reform advances scientific
learning for limited English proficient middle school students (Educational Practice
Report No. 17). Santa Cruz, CA: National Center for Research on Cultural
Montecel, M. R., and Cortez, J. D. 2002. Successful bilingual education programs:
Development and the dissemination of criteria to identify promising and exem-
plary practices in bilingual education at the national level. Bilingual Research
Journal 26, 1–22.
Nielsen, F., and Lerner, S. 1986. Language skills and school achievement of bilingual
Hispanics. Social Science Research 15, 209–40.
Pilkington, C. L., Piersel, W. C., and Ponterotto, J. G. 1988. Home language as a pre-
dictor of first-grade achievement for Anglo- and Mexican-American children.
Contemporary Educational Psychology 13, 1–14.
Ramirez, J. D. 1992. Longitudinal study of structured English immersion strategy,
early-exit and late-exit transitional bilingual education program for language-
minority children (Executive Summary). Bilingual Research Journal 16(1–2),
1–62.
Ramirez, J. D., Yuen, S. D., Ramey, D. R. 1991. Longitudinal study of structured English
immersion strategy, early-exit and late-exit transitional bilingual education programs
P1: JZZ
for language-minority children (Final report to the U.S. Department of Education).

San Mateo, CA: Aguirre International.
Reyes, P., Scribner, J. D., and Paredes Scribner, A. (eds.). 1999. Lessons from high-
performing Hispanic schools: Creating learning communities. New York: Teachers
College Press.
Rosnow, R. L., Rosenthal, R., and Rubin, D. B. 2000. Computing contrasts, effect
sizes, and counternulls on other people’s published data: General procedures
for research consumers. Psychological Methods 1, 331–40.
Royer, J. M., and Carlo, M. S. 1991. Assessing the language acquisition progress of
limited English proficient students: Problems and a new alternative. Applied
Rumbaut, R., and Portes, A. 2001. Ethnicities: Children of immigrants in America.
Berkeley, CA: University of California Press.
Rumberger, R. W., and Larson, K. A. 1998. Toward explaining differences in edu-
cational achievement among Mexican-American language-minority students.
Sociology of Education 71(1), 68–92.
Saldate, M., Shitala, S. P., and Medina, M. 1985. Bilingual instruction and academic
achievement: A longitudinal study. Journal of Instructional Psychology 12(1),
24–30.
Shepard, L., Taylor, G., and Betebenner, D. 1998. Inclusion of limited-English-proficient
students in Rhode Island’s grade 4 mathematics performance assessment (CSE Tech-
nical Report 486). Los Angeles: University of California, National Center for
Research on Evaluation, Standards, and Student Testing.
Short, D. J. 1994. The challenge of social studies for limited-English-proficient stu-
dents. Social Education 58(1), 36–8.
Slavin, R. E., and Calderón, M. 2001. Effective programs for Latino students. Mahwah,
NJ: Lawrence Erlbaum Associates.
Solano-Flores, G., and Trumbull, E. 2003. Examining language in context: The need
for new research and practice paradigms in the testing of English-language
learners. Educational Researcher 32(2), 3–13.
Stevens, R. A., Butler, F. A., Castellon-Wellington, M. 2000. Academic language
and content assessment: Measuring the progress of English language learners
(ELLs) (CSE Technical Report 552). Los Angeles: University of Califor-
nia, National Center for Research on Evaluation, Standards, and Student
Testing.
Stoops Verplaetse, L. 1998. How content teachers interact with English language
learners. TESOL Journal 7(5), 24–8.
Teddle, C., and Reynolds, D. 2000. The international handbook of school effectiveness
research. London: Falmer Press.
Tellez, K. 1998. Class placement of elementary school emerging bilingual students.
Bilingual Research Journal 22(2–4), 279–95.
Thomas, W. P., and Collier, V. P. 2002. A national study of school effectiveness for lan-
guage minority students’ long-term academic achievement. Santa Cruz, CA: Center
Tikunoff, W. 1985. Applying significant bilingual instructional features in the classroom.
Rosslyn, VA: National Clearinghouse for Bilingual Education (ERIC Document
Reproduction Service No. ED 338 106).
P1: JZZ
Troike, R. C. 1978. Research evidence for the effectiveness of bilingual education.

NABE Journal (3), 13–24.
U.S. Department of Education. 1994. Improving America’s Schools Act. Washington,
DC: U.S. Government Printing Office.
1998. Turning around low-performing schools: A guide for state and local leaders. Wash-
ington, DC: Author.
2001. No Child Left Behind Act. Washington, DC: U.S. Government Printing
Office.
Valdez-Pierce, L., O’Malley, J. M. 1992. Performance and portfolio assessment for lan-
guage minority students. NCBE Program Information Guide Series. Washington,
DC: National Clearinghouse for Bilingual Education.
Velez, W., and Fernandez, R. R. 1991. The ASPIRA Association school retention
study. Equity and Excellence 25(1), 55–61.
Wenglinsky, H. 2000. How teaching matters: Bringing the classroom back into discussions
of teacher quality. Princeton, NJ: Educational Testing Service.
Willig, A. C. 1985. A meta-analysis of selected studies on the effectiveness of bilin-
gual education. Review of Educational Research 55(3), 269–318.
P1: JZZ
0521859751c05a
212
table a.5.1. Comparative Evaluation Studies
Sample Research Outcome

Authors Characteristics Grades Program Design Measuresa Resultsb,c
Alanis Hispanic, low 5 Two-way Btwn group: – TAAS: Eng, Read, School 1:
(2000) SES, Participate (50:50) – TR: ELL TWI and Math (type of Read and Math: 2-way < non-ELL in
= 3 yrs (n = 56) score not clear) EO
CB989B/Genesee
– CO: Non-ELL School 2:

EO (n = 80) Read: TWI ELL > non-ELL in EO
– Norms class
Math: TWI ELL > non-ELL in EO
class
0 521 85975 1
Burnham- 1–12 Bilingual Btwn group: – CTBS: Read, Lang, Grade 5:

Massey (type not – TR: Bil (n = 115) and Math (NCEs) – Students in Bil scored at/above
and specified) – CO: Non-ELL – High school grade-level (NCEs = 48–52)
Piña EO (n = 492) proficiency – Grades 1–5, grow 12–16 NCEs
(1990) – Norms measures: Read, – Read, Lang, Math: Bil = EO
– Longitudinal Write, Eng, and Grades 7–8:
Math – Read: Bil = EO (NCEs = 42–45)
– No statistical – Lang: Bil (NCEs = 48–53) = EO
analyses run (NCEs = 47–48)
– Math: Bil (NCEs = 56–57) = EO
(NCEs = 52–54)
Grade 12:
– GPA and high school proficiency
tests: All students in Bil program
November 12, 2005
passed high school prof. measures

in Read, Write, Eng, and Math
– At grade level
16:50
P1: JZZ
0521859751c05a
Cazabon Hispanic, low 1–3 Two-way Btwn group: – Student attitudes Grade 3 in most recent cohort:
et al. SES, matched (50:50) – TR: TWI – District test of lang – Eng Read: TWI (2.9) = TBE (2.1)
(1993) on nonverbal (n = 76) skills – Eng Math: TWI (3.9) = TBE (2.7)
intelligence – CO: TBE – CAT grade – Span Read: TWI (3.1) = TBE (3.6)
(n = 70) equivalent (GE) – Span Math: TWI (3.3) = TBE (3.0)
Longitudinal scores in Eng Read – TWI slightly above grade level and
CB989B/Genesee
and Math TBE below grade level in Eng read

– CTBS Español: and math
Read and Math – Positive attitudes
Cazabon Hispanic, 4–8 Two-way Btwn group: – CAT grade – Read: TWI = EO Grades 4–8
et al. low-SES (50:50) – TR: TWI equivalent: Eng, – Math: TWI > EO Grades 4–5∗∗∗ ;
(1998) (n = 250) Read and Math TWI = EO Grades 6–8
0 521 85975 1
– CO: Non-ELL SABE grade – Span Read: TWI = TBE Grades 4–7
EO (n = 212) equivalent: – Span Math: TWI = TBE Grades 4,
– Longitudinal Spanish, Read and 6–7; TWI > TBE Grade 5∗∗∗
Math Grade 8:
– ANCOVA (Raven – At grade level in Eng read and
as covariate) math
– Below grade level in Sp read and
math
Collier Hispanic, low SES K-6 Bilingual Btwn group: – CTBS: Eng Read After 2 years (K-1):
(1992) (early and – TR1 = Early exit and Math (NCE – No difference by program
Re- late) – TR2 = Late exit gains over time) Upper elem grades, late-exit students:
analysis – n = Subset of – No statistical –Scored at 51st NCE in English math
of Ramirez’s 2,000 analyses reported and 45th NCE in English reading
November 12, 2005
Ramirez students – Reports other groups scored lower,

(1992) – Norms but no scores provided
16:50
(continued)
213
P1: JZZ
0521859751c05a
214
Curiel Hispanic, low 7–10 Bilingual Btwn group: – GPA Dropout rate:
(1986) SES, Matched: (type not – TR: Bil (n = 86) – Attendance – Bil (24%) < non-Bil (43%)∗∗
Gender, family specified) – CO: non-ELL – Retention Retention:
CB989B/Genesee
structure, EO (n = 90) – High school – Bil (5%) < non-Bil (15%)∗

parent graduation – Students in Bil education had
birthplace and significantly higher GPAs and
education lower school drop-outs the longer
they had participated in the Bil
program
0 521 85975 1
de la Hispanic, low 1–3 Bilingual Btwn group: – SAT: Eng By Grade 3:

Garza SES, Program (type not – TR: Bil (n = 24) – CAT: Eng – Students at grade level (T scores =
and partic = 3 yrs specified) – CO: non-ELL – CTBS Español: 48–53)
Medina EO (n = 118) Read and Math – Read voc: Bil > EO∗
(1985) – Norms – Raw scores – Read: Bil = EO
– Longitudinal transformed to – Math computation: Bil = EO
T-scores – Math concepts: Bil = EO
– Grade level in Eng and Span, Read
and Math
Fulton- Hispanic, low SES 1, 6 Bilingual Btwn group: – CTBS By Grade 6:
Scott (type not – Bil: (n = 20) grade-equivalent: – CTBS: Bil (6.9) > Non-int (4.9)∗
and specified) – Integrated ESL: Eng (assume total Bil (6.9) > Integ (6.0)∗
Calvin (n = 20) score, but not clear) Integ (6.0) = Non-int (4.9)
November 12, 2005
(1983) – Non-integrated – Grade point – Read GPA: Bil (4.7) > Non-int (2.9)∗
ESL: (n = 20) average: Read, Bil (4.7) > Integ (3.5) ∗
Lang, and Math Integ (3.5) = Non-int (2.9)
16:50
P1: JZZ
0521859751c05a
– Lang GPA: Bil (4.6) > Non-int (2.9)∗

Bil (4.6) > Integ (3.7)∗
Integ (3.7) = Non-int (2.9)
– Math GPA: Bil (4.7) > Non-int (2.8)∗
Bil (4.7) = Integ (4.1)
Integ (4.1) > Non-int (2.8)∗
CB989B/Genesee
– Bilingual at grade level

Gersten and Hispanic, low 4–7 Transitional Btwn group: – ITBS NCE: Eng, Grades 4–6:
Woodward SES, Program (early exit) – TR: TBE read, lang, and – TBE < SEI
(1995) partic = 4 yrs and (n = 117) vocab Grade 6:
Structured – CO: SEI (n = 111) – OLDM (Eng score – Both groups scored low:
English – Longitudinal used as covariate) – Lang Bil (39.2) < SEI (43.2)∗
0 521 85975 1
Immersion – Rate of entry into – Vocab Bil (25.9) = SEI (27.7)

(30–90 mainstream classes – Read Bil (32.8) = SEI (33.6)
min/day of – Attitudes
Span)
Grade 7:
– SEI = Bil
– Lang Bil (43.2) < SEI (44.4)
– Vocab Bil (33.5) = SEI (34.7)
– Read Bil (27.9) = SEI (28.6)
– SEI students significantly more
likely to enter mainstream (99%)
than Bil (65%)∗∗∗
– No difference in students’ attitudes
November 12, 2005
– By Grade 7, students scored low in

Vocabulary and Read, close to
grade level in Lang
16:50
215
(continued)
P1: JZZ
0521859751c05a
216

Lindholm Hispanic, 2–3 Two-way Btwn group: – CTBS NCE: Eng, By Grade 3:
CB989B/Genesee
(1991) low-SES (90:10) – TR: TWI Read and Math – ELLs scored low in Eng Read
(n = 168) All – T-tests & ANOVAs: (NCE = 27)
ELLs Determine group – Slightly below average in Eng Math
– CO: TBE differences in Read (NCE = 42)
(n = 118) and Math – Below average (NCE = 42) in Span
Read
0 521 85975 1
– Above average in Span Math (NCE

= 60)
– Comparing ELLs in TWI vs. TBE–no
significant differences in Grade 3
Lindholm- Hispanic, low 9–12 Two-way Btwn group: – Attitudes – Positive attitudes related to school
Leary SES (90:10) – TR: TWI grads, – GPA and academic competence;
and previously in – Enrollment in Math motivated; college prep behaviors;
Borsato TWI in courses enrolled in higher-level math
(2001) elementary – ANOVAs & χ 2 : courses (TWI students)
(n = 142) determine group – Comparing non-TWI to
– CO: non-TWI differences TWI students, TWI students like
(n = 17) school more∗∗ are more prepared
for 4-year college∗ : know more
November 12, 2005
about entrance requirements∗∗ ,

attend college info∗ , and enrolled in
higher-level math courses∗
16:50
P1: JZZ
0521859751c05a
Medrano Hispanic, low SES 7–9 Bilingual Btwn group: – CTBS scaled scores: Grade 6:
(1986) (type not – TR: BE (n = 179) Eng – Read: Bil (691) = EO (697)
specified) – CO: non-ELL EO – Math: Bil (704) > EO (699)∗
(n = 108) – Grade level in Eng Read and Math
– Retrospective
Ramirez Hispanic, low SES K-4 or Transitional – n = 2,000 – CTBS: Eng Read, – After 2 years, students do equally
(1992) K-6 (early exit students, Lang Arts, and well in Eng Read and Math
CB989B/Genesee
Large and late –Longitudinal Math – Bilingual education helps ELLs

Congress- exit) and (Students not – Trajectory Analysis catch up to Eng-speaking peers in
mandated Structured matched, but of Matched Eng Read, Lang Arts, and Math
study English programs Percentile (TAMP): – ELLs in Eng mainstream fall
Immersion matched for key comparison of behind
programmatic growth to norming
0 521 85975 1
characteristics) pop.
Saldate Matched- 1–3 Bilingual Btwn group: – WRAT: Eng and Grade 3:
et al. Hispanic, low (type not – TR: Bil (n = 31) Span for Grade 3 – Eng Read: Bil > EO∗
(1985) SES, PPVT specified) – CO: non-EL EO (type of score – Eng Spell: Bil = EO
(n = 31) unknown) – Eng Math: Bil > EO∗
– Longitudinal – Span Read: Bil > EO∗
– Span Spell: Bil > EO∗
– Span Math: Bil > EO∗
Thomas SITE #1 Rural: 4–7 French/ Btwn group: – Various tests (ITBS, BI vs MainEO:
and Mostly English – TR: ELL CTBS, Stanford 9, – Read BI (56) > MainEO (48)
Collier Franco-Am, low 1-way Bilingual Terra Nova) NCE: – Lang BI (59) > MainEO (51)
(2002) SES, ELLs have Immersion / BI Eng Read and – Math BI (55) > MainEO (52)
some academic (n = 47) Math 4 NCEs gps – Grade level in Eng Read and Math
November 12, 2005
prof in Eng – CO: ELL is sign, 4 small btw,

MainEO (n = 39) 6 moderate,
– Longitudinal 10 large
16:50
217
(continued)
P1: JZZ
0521859751c05a
218
CB989B/Genesee

Thomas SITE #2 Large 2–11 Transitional Btwn group: – Stanford 9 NCE: Grade 5 with min 4 years in program-
and Urban: (early exit) – TR: ELL TBE Eng, Read and longitudinal:
Collier Hispanic, low and ESL Math – Eng Read: TWI (51)> DBE (41) and
0 521 85975 1
(n = 3,333)
(2002) SES – CO: ELL ESL – APRENDA 2 NCE: TBE (41)
(n = 3,655) Spanish, Read and DBE (41) = TBE (41)
– CO: ELL Math – Eng Math: TWI (59)> DBE (51) and
MainEO – Drop-out rate TBE (50)
(n = 1,599) 4 NCEs between DBE (51) = TBE (50)
– CO: EO MainEO groups is – Span Read: TWI (65)> DBE (56)
(n = 103,887) significant, 4 NCEs and TBE (57)
small, 6 NCEs DBE (56) = TBE (57)
moderate, 10 NCEs – Span Math: TWI (63)> DBE (54)
very large and TBE (56)
DBE (54) = TBE (56)
Grade 11 – Cross-sectional:
– Read: EO (49) = TBE (47) > ESL
November 12, 2005
(40) > MainEO (25)

– Math: EO (46) = TBE (44) >
MainEO (34)
16:50
P1: JZZ
0521859751c05a
ESL (41) > MainEO (34)

Dropout rate: Bil/ESL and
EO (3%) < MainEO (4.6%)
– TBE close to grade level in Read
and Math
– ESL slightly below grade level in
CB989B/Genesee
both Read and Math

– Main EO very low in Read and low
in Math
Thomas SITE #3 Mid – 8–11 ESL Content – TR: ESL Content – ITBS NCEs Grade 8:
and sized urban: Instruction (n = 141) – Read and Write (NCE = 36–38)
Collier Hisp, low SES – Longitudinal – Math and Science (NCE = 41–49)
0 521 85975 1
(2002) Grade 11:

– Read and Write (NCE = 37–47)
– Math and Science (NCE = 44–48)
– Gr 11, below grade in Read, near
grade other content
a Uses norm-referenced standardized achievement test unless noted otherwise.
b The =, >, and < signs and probabilities (∗ p < 0.05, ∗∗ p < 0.01; ∗∗∗ p < 0.001) note the original determination of group differences; = is used if group
differences are not significant in the original statistical analyses.
c If there are multiple years of data provided in the research article, the results are presented for the most recent data.
November 12, 2005
16:50
219
P1: JZZ
table a.5.2. Descriptive Studies of Two-Way Immersion Programs

0521859751c05a
220
Sample Research
Authors Characteristics Grades Program Design Outcome Measures Results
De Jong Hispanic, low 5 Two-way Within group: – SAT NCE: Eng, Read and By Grade 5:
(2002) SES (50:50) – n = 95 Math – Students (above) average in Math
– Descriptive – Mass. State Curriculum (NCEs = 43–73)
Content Frameworks – Below grade level in Read (NCEs
CB989B/Genesee
(scaled scores) = 38–43)

– Students passed state measure of
proficiency in Lang Arts,
Mathematics, Science and
technology
Kirk- Hispanic, low 3–8 Two-way Within group: – Illinois Standards – Read and Math Grade 3–8: At
0 521 85975 1
Senesac SES, Partip = (80:20) – n = 24 Achievement Test (ISAT): grade level in Read & Math
(2002) min 5 years – Longitudinal Read, Write, Math, Social – On state assessments, most
– Descriptive Science, and Science students (81–88%) met the
– Illinois Goal Assessment standards in Read and Write; 2/3
Program (IGAP): Read, in Math and Science; 1/2 in Social
Write, Math, Social Science
Science, and Science – Pass rates significantly higher
– GE scores than those for the district and
state for ELLs
Lambert Hispanic, low 4–6 Two-way Within group: – Student attitudes – Most students have positive
and SES (50:50) – n = 32 – CAT grade equivalent attitudes toward school, TWI
Cazabon – Longitudinal scores: Eng, Read program, and other languages
(1994) – Descriptive – Raven Progressive and cultures
November 12, 2005
Matrices test of non- By Grade 6:

verbal abstract reasoning – Students at grade level in Read
– Average in a measure of abstract
16:50
reasoning
P1: JZZ
0521859751c05a
Lindholm Hispanic, low K-1 Two-way Within group: – CTBS NCE: Eng scores Grade 1:
(1988) SES (90:10) – n = 73 – Student attitudes – Slightly below average in Read
– Descriptive and Math
– Expected levels of academic
competence (attitudes), similar to
other Eng speakers
CB989B/Genesee
Lindholm Hispanic, low 1–4 Two-way Within group: – CTBS-U NCE: Eng Grades 1–2:
and SES (90:10 – n = 249 – Very low in Read (NCEs of 16–31)
Aclan and students – Moderately low in Math (NCEs of
(1991) 50:50) (n = 159 ELL) 26–44)
in 2 schools By Grade 4:
– Correlational – Almost average in Read (NCEs of
0 521 85975 1
42–53) and Math (NCEs of 46–54)

– Students with high levels of
bilingual proficiency scored at
grade level in Read and Math
(NCEs of 53–54)
Lindholm Hispanic, low K-6 Two-way Within group: – CTBS NCE: Eng Grade 6:
and SES (90:10) – n = 78 – Average in Math (NCE of 56)
Fairchild – Descriptive – Slightly below average in Read
(1989) – Longitudinal (NCE of 43)
(continued)
November 12, 2005
16:50
221
P1: JZZ
0521859751c05a
222
Sample Research
Authors Characteristics Grades Program Design Outcome Measures Results
CB989B/Genesee
Lindholm- Hispanic, low K-8 Two-way Within group: – Various norm-referenced Read & Lang:
Leary SES, urban, (90:10 – n = 8,000 standardized achievement– Students begin at 22nd NCE;
(2001) suburban and students in 18 tests: Eng, Read, Lang andmake significant progress (43rd
and rural 50:50) schools Math NCE)–not reach the 50th NCE by
– Descriptive Grade 7
and Math:
0 521 85975 1
correlational – Two-way students in 50:50 score

– Longitudinal average and those in 90:10
– Stats approach average by Grade 5
– Positive cross-cultural and lang
attitudes, and perceived scholastic
competence
Lindholm- Hispanic, low 9–12 Two-way Within group: – Student attitudes – Positive attitudes toward school,
Leary SES (90:10) – n = 142 – Drop-out rates want college degree, enrolled in
and – Descriptive – GPA college-bound Math courses, and
Borsato – Course placement in Math won’t drop out
(2001) – 42% felt TWI program kept them
Cited in from dropping out
Table 1
November 12, 2005
16:50
P1: JZZ
Conclusions and Future Directions
Fred Genesee, Kathryn Lindholm-Leary,

William M. Saunders, and Donna Christian
In Chapters 2 through 5, we presented research findings and methodolog-

ical issues specific to each domain of learning. In this chapter, we turn to
common trends in the research findings reviewed in Chapters 2 through 5
and then go on to identify directions for future research.
common trends
Role of ELLs’ First Language

The first notable trend is the influential role that ELLs’ native language
plays in their educational achievement. Maintenance and development of
ELLs’ L1 is influential in all domains we examined: oral language, literacy,
and academic achievement. The influence of the L1 was evident in stud-
ies that examined planned instructional or programmatic interventions
(Chapter 5) and those that examined unconscious, implicit processes that
are implicated in literacy and oral language development (in Chapters 2
and 3; for example, when ELLs draw on their knowledge of cognates in
the L1 when decoding words in the L2 or the transfer of reading compre-
hension strategies from the L1 to the L2). In citing evidence in support of
maintaining and using ELLs’ L1, we do not deny the critical importance
of English for educational achievement. We noted in Chapters 2 and 4 that
there is an important link between L2 exposure and proficiency and the
development of literacy skills in English. However, the importance of ELLs’
L1 raises an educational challenge. How can the developmental comple-
mentarity between L1 and L2 be used to their educational benefit at a time
when legislation in some states disfavors instruction through the medium
of languages other than English? Equally important, how can the L1 be
maintained when most ELLs, in those states and others, are in fact enrolled
in programs where English is the sole language of instruction? There is a
need to explore noninstructional uses of ELLs’ L1 in order to determine
223
P1: JZZ
224 Fred Genesee et al.
how it can continue to develop and benefit their literacy and academic
development. For example, the use of the L1 could be promoted in family
and community contexts by having parents read to children or children
read to parents in the L1; by providing after-school programs for students
where the L1 is used for literacy or academic purposes; or by ensuring
that L1 books and videos are stocked in public libraries. The effectiveness
of such extracurricular interventions warrants empirical investigation to
determine their impact on ELLs’ educational success.
Developmental Links between First and Second Language

A second major trend to emerge from this collection of studies concerns
the nature of the developmental links between L1 and L2 in the educa-
tion of ELLs. Under one hypothesis, it might be expected that high levels
of general L1 proficiency would be related to high levels of L2 oral lan-
guage, reading and writing, and academic achievement. However, a care-
ful reading of the relevant research indicates that specific aspects of L1
ability (e.g., understanding of paradigmatic word associations, the ability
to provide formal definitions of words, knowledge of L1–L2 cognates) are
more influential in L2 development, especially in domains related to L2
literacy and academic achievement. A similar pattern was found for the
developmental relationship between L2 oral ability and L2 literacy. We,
and others (e.g., Cummins, 2000) have characterized these critical aspects
of L1/L2 oral ability in various ways: sometimes as components of L1/L2
oral proficiency that are related to academic purposes; sometimes as oral
language that serves higher order cognitive processes; and sometimes as
thorough or deep understanding of language in contrast to surface-level
proficiency or understanding. The true nature of these critical features
of L1 and L2 oral proficiency is clearly an important direction for future
research. Although we lack complete understanding of the precise nature of
these skills, it is clear that educators should adopt differentiated and com-
plex views of L1 and L2 development and their interrelationship if they
are to optimize ELLs’ development of language for academic purposes.
We have sufficient knowledge at this time to devise strategic and focused
language teaching/learning interventions that could enhance the educa-
tional outcomes of ELLs.
Instructional Approaches
Third, interactive learning environments that provide carefully planned
direct instruction of target language skills, as needed, are likely to be
most effective. Moreover, language skills that are linked to literacy and
academic domains should be the target of such instruction. Home-based
interventions or bilingual instruction in school that focus on L1 language
P1: JZZ
Conclusions and Future Directions 225
development should emphasize language skills implicated in higher order

cognitive processes and language for literacy and academic purposes.
There is considerable transfer of these kinds of L1 language skills to the L2
and these are the skills that are associated with achievement in literacy and
academic domains. More specifically, for example, ELLs can benefit from
knowing that there are certain words in the L1 that have the same form as
words in the L2 and that some of these words also share meaning. Even in
English-only school settings, emphasis on English for academic purposes
is likely to have the greatest payoff in student achievement. An emphasis
on language for literacy and academic purposes, be it the L1 or L2, does
not mean that language skills for day-to-day communication should be
neglected. However, development of language skills for day-to-day com-
munication is insufficient to promote high levels of literacy and academic
achievement in school.
Subgroup Differences
A fourth finding to emerge from our analyses of this corpus of studies is the
significance of differences among subpopulations of ELLs. The importance
of differences between students was indicated in research on family and
community variables as well as in research on crosslinguistic and cross-
modal aspects of literacy development. Differences among subpopulations
of ELL students was evident in multiple ways – in their learning strate-
gies, their transfer of skills from the L1 to the L2, their motivations, and
the influence of L1 use at home on achievement in school, among others.
These kinds of differences were found in first versus later generations of
Hispanic ELLs, in ELLs from low versus those from more advantaged
socioeconomic backgrounds, and in ELLs with Hispanic versus Asian cul-
tural backgrounds. Other neglected but important sources of difference are
age or grade level upon entry to U.S. schools, capacity to learn (i.e., students
with typical capacities for learning versus those with impaired capacities),
and language typology. Student diversity warrants increased attention in
future research, especially at the current time when accountability, and
especially standards-based accountability, risks overshadowing pedagog-
ical issues linked to diversity.
Time
Finally, time emerged as a significant factor in student learning. There are
two aspects of time that are important: duration and consistency. The effects
of duration were most evident in program evaluation studies. Evaluations
of educational programs for ELLs who were in the program for a relatively
short period of time produced results, in some cases, that differed sharply
from evaluations that were conducted after the students had participated
P1: JZZ
in the program for some time. Later evaluations produced more positive
outcomes than early evaluations. It makes pedagogical and developmental
sense that time is important because learning takes time and is cumulative.
As a result, the true effects of educational programs are likely to be evident
only after some time.
Consistency, another dimension of time, matters. Consistency of educa-
tional experience affords students developmentally coherent opportunities
to learn. Programs that plan for coherence across grade levels and devel-
opmental stages yield more positive student outcomes than programs that
do not. In fact, evaluation studies have shown that students exposed to a
variety of different approaches perform poorly in comparison to students
who have had consistent exposure to the same program over time. This
is yet another reason why educators need comprehensive frameworks for
planning instruction over time, as noted previously. We return to a discus-
sion of the importance of development in planning research on ELLs in
the next section. Continuous, coherent, and developmentally appropriate
educational interventions are absolutely critical if ELLs are to achieve their
full potential in school.
future directions
There is an overarching need for sustained, programmatic research that
aims to build and test models of effective teaching and successful learning
in school settings with ELLs. This calls for research whose primary goal is
theory development. Shavelson and Towne (2002) have made a similar rec-
ommendation in the National Research Council report on Scientific Research
in Education (see “Scientific Principle 2: Link Research to Relevant Theory”;
Executive Summary, p. 7). Single studies with an immediate applied ori-
entation lack generalizability and, thus, applicability to other educational
settings. Widespread application of research findings to the benefit of large
numbers of ELLs is more likely to come from sustained research efforts
whose primary aim is a full and in-depth understanding of an issue than
from one or two isolated studies on a specific topic. Applied research con-
sisting of single studies is not as useful as theory-driven research in identi-
fying the needs of ELLs across the United States. Support for theory-driven
research with extended time frames calls for funding agencies and research
institutes to support research activities without an emphasis on immedi-
ate application. It also calls for political authorities, in their policies and
funding initiatives, to balance “improving education immediately” with
“expanding our understanding” so that education can be improved over
the long run. The latter requires time and material resources.
At the same time, consumers of educational research need to appreciate
that findings about “best practices” do not necessarily mean “single best
practice.” Policymakers and the public at large must come to understand
P1: JZZ
that there is not just one way to teach ELLs effectively. To the contrary, there
are alternative ways to achieve satisfactory oral language, reading and writ-
ing, and academic outcomes for ELLs. Indeed, it is highly unlikely that a
single instructional approach or method is likely to be effective for all ELLs
given the diversity of backgrounds, resources, and challenges they bring to
the learning environment, often within a single classroom. Educators need
to be able to make use of alternative instructional approaches to promote
the development of all ELLs under diverse and changing circumstances.
Indeed, it is likely that equal outcomes can be achieved through the effec-
tive application of alternative methods. The important point here from
a research perspective is that research designs that compare one method
against another may be useful to answer some types of questions, but other
designs may be needed when it is unreasonable to expect that there is one
best or single answer. The challenge facing educational researchers is to
identify the successful methods in terms of the circumstances and learners
for which they are successful. To do this, researchers must apply a diversity
of comparison points.
In general, the complexities of the educational enterprise call for varied
and multiple research designs, including case studies, ethnographies, and
classic experimental and quasi-experimental designs. In the field of mono-
lingual language acquisition, for example, case studies and ethnographies
have played critical roles in early theory development; Brown’s case stud-
ies of Adam, Eve, and Sara (Brown, 1976) and Heath’s ethnography of lan-
guage and literacy development in children from minority backgrounds
(Heath, 1983) are classic examples. Likewise, in the field of bilingualism,
Leopold’s monumental case study of his daughter, Hildegaard, was instru-
mental in launching research on bilingual acquisition and is still widely
cited for its findings (Leopold, 1949). The case study approach is particu-
larly useful during the initial stages of investigation into new issues or as a
way of developing new theories. At the same time, more controlled exper-
imental and quasi-experimental designs are essential if we are to expand
exploratory investigations in systematic ways, to acquire in-depth under-
standing of particular issues, and to test the generalizability of alterna-
tive theoretical possibilities. Experimental and quasi-experimental designs
are an important way of testing educational theory; they are required to
systematically examine, refine, and elaborate theoretical possibilities and
to test applications. In short, it is unlikely that a single research design
or methodology will be able to address the full complexity and range of
issues faced by educators working with ELLs. As Shavelson and Towne
(2002) note: “. . . scientific claims are significantly strengthened when they
are subject to testing by multiple methods. . . . Particular research designs
and methods are suited for specific kinds of investigations and questions
but can rarely illuminate all the questions and issues in a line of inquiry”
(p. 7, Executive Summary).
P1: JZZ
We offer suggestions for future research in the following sections under

three main headings: (1) Developmental Research, (2) Learners, and (3)
Classrooms. We end with a recommendation for follow-up state-of-the-art
reviews.
Developmental Research
Research on the oral language, literacy, and academic achievement of ELLs
is fundamentally about learners’ “development” in educational settings –
how student competencies change over time and the maturational, socio-
cultural, and pedagogical factors that facilitate or impede change. Truly
longitudinal research on teaching and learning in settings with ELLs would
be of significant benefit to the education community because it would yield
insights about the complexities of the educational development of ELLs.
Much of the research reviewed in this volume has examined learners at a
single grade level or different learners at several grade levels (see, how-
ever, Howard, Christian, and Genesee, 2003; Reese et al., 2000). As a result,
we have scant understanding of the actual developmental changes that
ELLs go through during the acquisition of oral language, reading and writ-
ing, and academic skills from beginning level to mature, advanced levels.
Research that focuses on ELLs at specific grades can give the impression
that what is true for one age group is equally true for another and what
works at one stage of development works at another. We need longitudi-
nal research designs to test the extent to which this is really true. Lacking
solid longitudinal research, we risk exposing students at different stages
of development to ineffective learning environments.
Investigating the developmental changes that the same learners go
through from grade to grade would contribute to our understanding of
the role of specific maturational, sociocultural, and pedagogical influences
on achievement and how these change and interact as learners mature
and engage in school and community life. Since development is dynamic,
a different combination of learner and instructional factors will probably
account for learning at different grade levels. For example, it is likely that
different components of oral and written language play important roles
in learning to read and write at different stages of acquisition. We know
from research on monolingual literacy acquisition that while phonologi-
cal awareness plays a significant facilitating role in early stages of read-
ing acquisition and, in particular, decoding, its importance diminishes in
later stages. Likewise, when it comes to instruction, it is likely that specific
instructional approaches and/or techniques will be differentially benefi-
cial when learners are beginning to learn to read and write in comparison to
when they have attained relatively advanced levels of skill at reading and
writing. An understanding of these developmental changes can provide a
solid basis for curriculum design and instructional intervention. There is
P1: JZZ
a critical need for research on the development of writing skills. This was
by far the most neglected domain of development and yet is, arguably, the
most challenging since it draws on all other skills – oral language, reading,
academic, and cognitive.
In the same vein, future research on the co-development of oral lan-
guage, literacy, and academic skills is critical if we are to understand the
developmental interdependencies of these interrelated skills and if we are
to design educational initiatives that facilitate their co-development. There
is a particularly strong need for research that examines the links between
oral language and literacy development on the one hand and between
oral language development and academic achievement on the other hand.
We also need to know whether the same interdependencies in oral and
written language characterize achievement in different academic subjects.
We have much to learn about the level and kinds of oral language and
literacy skills necessary to promote grade-appropriate academic achieve-
ment by ELLs and whether these depend on the nature of the subject
matter – mathematics versus science versus social studies, for example.
We also need to understand better the differential role of oral language
and literacy (whether in L1 or L2) in fostering academic achievement at
different grade levels as academic subject matter becomes more abstract,
complex, and arguably language-dependent. This is an especially impor-
tant issue in the education of ELLs who enter American schools in middle
or high school and particularly in the case of students who have not had
the benefit of prior education.
There are challenges to conducting developmental research in real
school settings. This can be illustrated in the program of research initiated
by Saunders and Goldenberg, a CREDE study designed to test, over time,
the independent and combined effects of several instructional components
within a successful transitional bilingual education program in southern
California (see Saunders, 1999). During the study, legislation was passed in
California that led to the demise of many bilingual programs so that the pro-
grams targeted in the study changed. While some of the intended research
was completed (see Saunders and Goldenberg, 1999; and Saunders and
Goldenberg, 2001), the longitudinal aspect of the project, although critical
for investigating the success of students’ transition, was not completed.
Thus, while researchers and funding agencies should plan for sustained,
longitudinal research, they should also be prepared for disruptions in their
efforts that are brought about by circumstances beyond their control.
Learners
As we noted earlier, the lion’s share of research attention has been on
ELLs from Hispanic backgrounds. There is a need for research on the
development of learners from other major ethnolinguistic groups in the
P1: JZZ
United States. Students of Vietnamese, Hmong, Cantonese, and Korean

background need to be examined because they are the next most populous
groups of minority students in the United States (Kindler, 2002). Includ-
ing different ethnolinguistic groups is particularly important in research
on the influence of instructional and noninstructional factors in order to
determine if the same constellation of instructional and family/community
influences accounts for learning when students come from different SES
and language backgrounds. Research by Steinberg, Dornbusch, and Brown
(1992) has shown that the particular constellation of factors that accounts
for successful learning is different for Asian American, African American,
and Hispanic high school students. Replication and extension of this line
of research, including factors that are pertinent to ELLs specifically, would
be useful. There is also a pressing need for additional research on ELLs
in higher grades and on ELLs who enter the U.S. educational system in
middle or high school, particularly those with little or no prior schooling.
The learning demands on these students are especially challenging, and
educators need more research on these particular students if they are to
respond effectively to their needs.
We need to expand our understanding of English oral-language devel-
opment in ELLs. English is both a goal of educational instruction and a
means to achieving other goals and, in particular, academic excellence.
Yet, of all the research domains we examined, oral-language development
received the least empirical attention. We need research that describes the
developmental course of oral-language proficiency in ELLs from different
language/cultural backgrounds and in ELLs who enter the United States
at different ages with different amounts of prior education and exposure
to English. Moreover, as noted previously, we need empirical evidence
of the relationship between oral-language development and academic
achievement; for example, what oral-language skills are needed to suc-
ceed in mastering academic subjects and how do these change over time?
Additional research on ELLs with impaired capacities for language
and/or academic learning is also needed if we are to address the needs
of all ELLs, those with typical ability to learn as well as those with various
disabilities. Future research would benefit from more detailed documen-
tation of students’ specific impairments. In this regard, researchers must
be careful to differentiate students with endogenous impairments from
those who are simply delayed in their language learning and/or academic
achievement because of their second-language status. Current published
research has shown little sensitivity to these confounding possibilities.
There is also a serious need for research on the incidence and nature of
impairment among ELLs and on the validity of assessment methods used
to identify such students. Future research into the nature and extent of
impairment in ELLs in comparison to monolingual English-speaking learn-
ers would be particularly useful because it would indicate whether ELLs
P1: JZZ
who have learning/language impairments suffer from the same kinds of

impairments and to the same extent as monolingual learners. This, in turn,
would have considerable implications for the identification and education
of ELLs with atypical capacities to learn. Regardless of ELLs’ particular
ethnolinguistic group membership or learning capacity, future research on
ELLs must provide much greater detail about their linguistic, educational,
and family backgrounds.
It would also be useful to conduct research on different groups of ELLs
in the same studies. This would make it possible to ascertain how the oral
language, literacy, and academic development of various groups differ
and this, in turn, would help us to begin to identify which instructional
interventions are appropriate in each case. For example, students may be
(1) truly monolingual ELLs, (2) have relatively advanced levels of profi-
ciency in both the L1 and the L2, (3) have stronger L1 than L2 skills, or (4)
vice versa, have stronger L2 than L1 skills. A fundamental question within
this domain of research is how the linguistic, cognitive, and educational
development of students is influenced by their status as monolingual learn-
ers, proficient bilingual learners, or second-language learners. Addressing
this fundamental question requires careful and detailed information about
ELLs’ language status.
Classrooms
The recommendations for future research in this section focus on classroom
contexts. Our review revealed considerable research on alternative instruc-
tional approaches/strategies for teaching literacy to ELLs and a number
of important general conclusions emerged from that review. As we noted
in Chapter 4, educators need more than an array of specific methods or
activities that they can draw on when planning literacy or academic sub-
jects. They need comprehensive frameworks for selecting, sequencing, and
delivering instruction over the course of an entire year and from grade to
grade. Two frameworks that provide such guidance are the Five Standards
for Effective Pedagogy (Tharp et al., 2000) and the SIOP model for integrat-
ing language and content instruction (Echevarria, Vogt, and Short, 2000).
Taken together, these two frameworks provide tools for planning education
that can incorporate a variety of instructional approaches to ensure that the
learning environment is meaningful, coherent, and individualized. While
both frameworks enjoy some empirical support (see Echevarria, Short, and
Powers, 2003; and Tharp et al., 2000), extension of this work would serve
to expand our understanding of the scope of their effectiveness. Neither
framework is prescriptive but lends itself to variation and modification.
Future research on specific variations of the Five Standards and SIOP mod-
els, alone or in combination, with students from different language back-
grounds, at different ages/stages of development, and for different subject
P1: JZZ
matter would help us refine our understanding of the conditions under

which they work well.
Classrooms vary significantly from one another with respect to num-
ber of students, language and cultural backgrounds of students, SES, and
prior literacy training, to mention some obvious dimensions of variation.
Moreover, classrooms with ELLs often change as students enter and leave.
With the exception of Lindholm’s research on the effectiveness of two-
way immersion programs in classrooms with different ethnolinguistic
compositions (Lindholm-Leary, 2001), we have little understanding of how
classroom composition affects teaching and learning or of how teachers
cope with classrooms with different compositions of students. Some press-
ing questions arise: What is the nature of oral-language use, both teachers’
and students’, in classrooms with students whose mastery of oral English is
limited? How can the oral-language development of ELLs be promoted in
classrooms with learners with varied levels of English proficiency? How
can instruction through English be delivered to ensure comprehension
and mastery of academic material in classrooms with learners who have
different learning needs? How does change in student composition over
the course of the school year influence student interactions and learning?
Future research is called for that focuses on the classroom as the unit of
analysis in order to better understand the social and intellectual dynamics
of classrooms and how to design instruction that is effective in different
classroom contexts.
In a related vein, classrooms and the schools in which they are located
do not exist in a vacuum. They are part of larger, more complex, and chang-
ing communities. Educators often remark on the relationship between the
school and the community and the efforts they make to bring about col-
laboration between schools and communities. Future research with the
community as the unit of analysis would help move us beyond impression-
istic speculation to empirically grounded knowledge. Some fundamental
questions arise: Does the ethnolinguistic composition of the community
affect the expectations and goals of the community vis-à-vis their chil-
dren’s education and, by extension, the role of their children’s teachers?
What can or should teachers do in response to communities that might
differ from mainstream Anglo-American communities? What can and
should communities do to facilitate their child’s education in mainstream
schools?
While issues concerning teachers and professional development
are dealt with in volume (details in this), we also believe that
attention needs to be paid to teachers, including their levels and kinds of
professional development, their understanding of different instructional
and assessment approaches, their knowledge and application of second-
language acquisition theory, and the processes that are required to ensure
that new teachers acquire competence in using new approaches.
P1: JZZ
Future Reviews
In closing, our final recommendation is that systematic reviews of research
findings on the oral language, reading and writing, and academic devel-
opment of ELLs be undertaken on a periodic and regular basis. This would
permit researchers and educators to take stock of current research on the
education of ELLs and of our progress in investigating issues critical to
planning effective education for these learners. In addition, there is a need
for considerably more support of research on the education of ELL stu-
dents. The statistics are clear – ELLs will constitute an ever-expanding and,
thus, important portion of the school-age population. Effective education
for ELLs means planning for their and the nation’s future.
References
Brown, R. 1976. A first language. Harmondsworth, UK: Penguin.
Echevarria, J., Short, D., and Powers, K. 2003. School Reform and Standards-Based
Education: How do Teachers Help English Language Learners? Technical Report.
Santa Cruz: Center for Research on Education, Diversity & Excellence.
Echevarria, J., Vogt, M. E., and Short, D. J. 2000. Making content comprehensible
for english language learners: The SIOP model. Needham Heights, MA: Allyn
& Bacon.
Heath, S. B. 1983. Ways with Words. New York: Cambridge University Press.
Howard, E. R., Christian, D., and Genesee, F. 2003. The development of bilingualism
and biliteracy from grade 3 to 5: A summary of findings from the CAL/CREDE study
of two-way immersion education. Santa Cruz, CA, and Washington, DC: Center
Kindler, A. L. 2002. Survey of the states’ limited English proficient students and avail-
able educational programs and services: 2000–2001 summary report. Washington,
DC: National Clearinghouse for English Language Acquisition & Language
Instruction Educational Programs.
Leopold, W. 1949. Speech development of a bilingual child: Volume 3. New York: AMS
Press.
Lindholm-Leary, K. J. 2001. Dual language education. Clevedon, UK: Multilingual
Matters.
Reese, L., Garnier, H., Gallimore, R., and Goldenberg, C. 2000. Longitudinal
analysis of the antecedents of emergent Spanish literacy and middle-school
tional Research Journal 37(3): 633–62.
Saunders, W. 1999. Improving literacy achievement for English learners in tran-
sitional bilingual programs. Educational Research and Evaluation 5(4), 345–
81.
Saunders, W., and Goldenberg, C. 1999. The effects of instructional conversations
and literature logs on limited- and fluent-English proficient students’ story
comprehension and thematic understanding. The Elementary School Journal
99(4), 277–301.
P1: JZZ
2001. Strengthening the transition in transitional bilingual education. In D. Chris-

tian and F. Genesee (eds.), Bilingual education. Alexandria, VA: Teachers of
English to Speakers of Other Languages, Inc.
Shavelson, R. J., and Towne, L. 2002. Scientific research in education. Committee
on Scientific Principles for Education Research. Washington, DC: National
Academies Press. http://www.nap.edu/execusumm/0309082919.html
Steinberg, L., Dornbusch, S. M., and Brown, B. B. 1992. Ethnic differences in ado-
lescent achievement. American Psychologist 47(6), 723–9.
Tharp, R. G., Estrada, P., Dalton, S. S., and Yamauchi, L. 2000. Teaching transformed:
Achieving excellence, fairness, inclusion and harmony. Boulder, CO: Westview
Press.
P1: JtR
0521859751apx CB989B/Genesee 0 521 85975 1 November 12, 2005 13:17
Appendix A
Definitions of Abbreviations in Research

Summary Tables
%ile percentile
ach achievement
AM American
assign assignment
bil bilingual
BINL Basic Inventory of Natural Language
BOLT Bahia Bay Oral Language Test
BR basal reader
BSM Bilingual Syntax Measure
btwn between
CAT California Achievement Tests
CO Control Group
cog cognitive
comp comprehension
coop cooperative
corr correlation
CTBS California Test of Basic Skills
def definition(s)
diff difference
dom dominant
SABE Spanish Assessment of Basic English
ELL English language learner
Eng English
EO English only
ESL English as a second language learner or program
FEP fully English proficient; fluent English proficient
FLS Functional Language Survey
GE grade equivalent (score)
gp(s) group(s)
GPA grade point average
235
P1: JtR
236 Appendix A: Definitions of Abbreviations in Research Summary Tables
gr grade
grads graduates
hi high
Hisp hispanic
IC Instructional Conversations
Integ Integrated classroom
IPT Idea Proficiency Test
ITBS Iowa Test of Basic Skills
K kindergarten
L1 first/native language
L2 second language
lang language
LAB Language Assessment Battery
LAS-O Language Assessment Scales-Oral
LEP limited English proficient
LD learning disabled
MainEO Mainstream English only classroom
MANOVA multivariate analysis of variance
Mass Massachusetts
Max maximum
med medium
multi multiple
n sample size
NCE normal curve equivalent
non-int non-integrated classroom
NSST Northwest Syntax Screening Test
obs observations
partic participation; participants
PIAT Peabody Individual Achievement Test
PPVT(-R) Peabody Picture Vocabulary Test (Revised)
pro pronoun
prof proficient/proficiency
pt point
r coefficient of correlation
rdg reading
rdg comp reading comprehension
RSQ Reading Strategy Questionnaire
SAT Stanford Achievement Tests
SE special education
SES socio-economic status
sig, sign significant/significantly
Sp, Span Spanish
SOLOM Student Oral Language Observation Matrix
SRA Science Research Associates
P1: JtR
Appendix A: Definitions of Abbreviations in Research Summary Tables 237
stats statistical
SW southwest
TAAS Texas Assessment of Academic Skills
TABE Texas Association for Bilingual Education
TR Treatment Group
TVIP-H Test de vocabulairio en Imagenese Peabody- Adaptación
Hispanoamericana
Voc/vocab vocabulary
wk week
WLPB-R Woodcock Language Proficiency Battery-Revised
WRAT Wide Range Achievement Test
wtg writing
yr year
P1: JtR
238
P1: irk
0521859751ind CB989B/Genesee 0 521 85975 1 November 12, 2005 14:37
Index
References followed by a “t” refer to tables.
Abedi, J., 197, 198, 199 Bacon, H., 128

Abella, R., 197 Bahia Bay Area Oral Language Test (BOLT),
Abramson, S., 67, 75, 132 35
academic achievement, 176–177 Baker, E., 199
assessment of, 196–200 Barolome, L., 64
family and background in, 193 Basic Inventory of Natural Language
future research on, 229 (BINL), 19, 34, 35
influence of language upon, 188–190 Bean, 79
instructional issues in, 190–192 Beeman, M., 17, 74
L2 oral language proficiency and, 29, Bermudez, A. B., 80, 114, 116
41 Betebenner, D., 200
length of U.S. residence and, 194–195 bilingual immersion instruction, 182
socioeconomic status and, 193–194 bilingual instruction, 5, 126, 127
academic language proficiency, 65–66 descriptive studies of, 185–186
accommodations, for standardized tests, effectiveness of, 177
198–200 ESL compared with, 180
Aclan, Z., 70–71, 189 ESL and SEI compared with, 181–183
Adams, D., 193 lack of consistency in, 201
Alanis, I., 179 studies of, 178–179
Alderman, D. L., 196 bilingualism, 188
Allen, V., 75 “early-exit” bilingual programs, 5,
Alvarado, C. G., 35 183–185
Asian Americans, 68, 142, 230 Bilingual Syntax Measure (BSM), 19, 34,
assessment 35
of academic achievement, 196–200 Blum, I. H., 120, 133–134
of literacy instruction, 136–139, 143 Borsato, Graciela, x, 180, 186
of oral language proficiency, 34–39, 41–42 Boscadin, C. K., 199
studies of, 34–61t Bousquet-Gutiérrez, M., 197
see also standardized tests Boyle, O., 71
attitudes of students, studies of, 186 Brown, B., 230
auditory discrimination, 115 Brown, R., 227
August, Diane, x, 28, 192 Buriel, R., 68, 78, 132, 135
Avila, E., 112 Burnham-Massey, L., 129, 179
239
P1: irk
240 Index
Butler, F. A., 197, 198 direct instruction, 110, 111–116, 139–140, 141
Butler, Y. G., 197 studies of, 149t–154t
Dixon, J. K., 192, 204
Cabello, B., 138 Doherty, R. W., 124, 141
Calderón, M., 128 Dornbusch, S. M., 230
Calvin, A. D., 128, 182, 187 drop-out rates, 1
Cancino, H., 16 Drum, P., 75
Cardoza, D., 34, 68, 78, 132, 135 dual language immersion programs, see
Carlisle, J., 17, 74 two-way immersion programs
Carlo, M. S., 71, 78, 198 Duran, B., 136
Castellon-Wellington, M., 197, 200 Duran, R. P., 194, 195
Cathart-Strong, R. L., 28 Durgunoglu, A., 73, 74, 75
Center for Research on Education, Diversity
& Excellence (CREDE), ix Echevarria, J., 115, 124, 140, 191
Chávez, R., 27 Edelsky, C., 80
Chesterfield, K. B., 19–20, 27 Ekboir, J. M., 193
Chesterfield, R. A., 19–20, 27, 29, 30 Elementary and Secondary Education Act
Chinese Americans, 120 (U.S.), 3
Christian, Donna, 129 English language development (ELD)
classroom context for learning, 231–233 programs, 5
cognate vocabulary, 75–76, 83 English (language)
cognitive development, 66 cognates for Spanish and, 75
cognitive learning strategies, 21 instruction in L1 versus, 126
Cohen, A. S., 118, 119 proficiency in, for academic work, 66
collaborative reading approach, 192 English as a Second Language (ESL)
college, 186 programs, 5
Collier, V. P., 78, 129, 132, 179, 180, 182, 183, bilingual instruction compared by, 180
194–195, 202, 204 bilingual instruction and SEI compared
Commins, N., 38 with, 181–183
communications, contexts of, 65 rates of oral language proficiency
community, 142, 232 development for, 26
literacy instruction and, 131–132 English immersion programs, 128
comparative evaluation studies, 177–181, “English-only” (EO) students, 11
212t English Peabody Picture Vocabulary Test
contexts of communications, 65 (EPPVT), 31
Contrastive Analysis Hypothesis, 66–67 environment
Corbett, C., 74 in interactive instruction, 116–117
Cronnell, B., 75 in school, 187
Cuevas, Gil, x
Cummins, J., 65–66 family, 142
Curiel, H., 178, 180 academic achievement and, 193
curriculum, 187 literacy instruction and, 131–132
literacy practices and, 133–134
Davidson, F., 197 oral language use in, 31
Davis, L., 17, 74 studies of influence of, 171t–175t
Davis, W. A., 123 use of L1 in home and, 134–136
definitions of words, 17 see also home
de la Luz Reyes, M., 124 Farest, C., 121
developmental bilingual programs, 5 Fashola, O., 75
developmental effects, 67 Fernández, M., 37–38
developmental interdependence Fernandez, R., 188–189, 193, 194
hypothesis, 65 Ferris, M., 130
P1: irk
Index 241
Field, M., 81 Idea Oral Language Proficiency Test (IPT-I),

Five Standards for Effective Pedagogy, 35
140–141, 231 Ima, K., 132, 136, 138
Fix, M., 2 immigrant students, 2
Fletcher, T. V., 35 academic achievement by, 194, 202
fluent English proficient (FEP) speakers, 189 Improving America’s Schools Act (U.S.,
Fradd, S. H., 198 1994), 196
Frontera, L., 137 Individualized Developmental English
Fulton-Scott, M. J., 128, 182, 187 Activities Placement Test (IDEA), 34
instruction
Galindo, R., 81 approaches to, 109–111
Gallimore, R., 78, 132 assessment of, 136–139
Garcia, G. E., 75, 138 direct, 111–116, 149–153
Garcia-Vázquez, E., 19, 190 influence on academic achievement of,
Garnier, H., 78, 132 190–192
Genesee, Fred, 3, 129 interactive, 116–121, 155t–161t, 224–225
genres, 80–81 language of, 126–131
Gersten, R., 128, 180, 182–183, 201 language of, studies of, 166t–170t
Glass, G., 36 in literacy, 109
Goldenberg, C., 78, 118, 119–120, 132, 141, methodological considerations in,
229 124–126
Goldstein, B., 17, 71 process approaches to, 121–124, 162t–165t
Gomez, L., 39 interactive instruction, 110, 116–121,
Gomez, R., 39, 123 139–140, 141, 224–225
Gonzalez, P., 16 environments for, 117t
Graham, 79 studies of, 155t–161t
interlanguage theory, 67
Hakuta, K., 26–27, 192, 197
Hamahan, Else, x Jackson, N. E., 133
Hampton, S., 193 Jacob, E., 28
Hancin-Bhatt, B., 73, 74, 75 Jilbert, K., 80
Hansen, D. A., 32–33, 135 Jimenez, R. T., 75, 78, 79, 80, 81
Harris, K., 71 Johnson, D. M., 29, 137
Hayes-Latimer, K., 27
Heath, S. B., 227 Kang, S., 75
Hernández, J. S., 114–115 Kennedy, E., 68, 132, 135
Hispanics Klein, M., 71
Asian Americans compared with, 68 Klingner, J., 110, 192
bilingual, academic achievement of, 188 Kramer, V. R., 115
language differences among, 37 Kucer, S. B., 115, 124
Hispanic students, 2
Hoffman, J., 121 “L1” (native language), 11
Hofstetter, C., 197, 199 academic achievement and, 189
home assessment of oral language proficiency
oral language use in, 31 in, 34
use of L1 in, 68, 134–136 concepts lexicalized in, 38
see also family crosslinguistic relationships between L2
Hopkins, K. D., 123, 124 and, 64–65
Horowitz, R., 137 developmental interdependence
Howard, E. R., 129 hypothesis on relationship between L2
Hsia, S., 75 and, 65
Hughes, M. T., 134 developmental links between L2 and, 224
P1: irk
242 Index
“L1” (native language) (cont.) Language, development, studies of, 46t–51t

importance of maintenance of, 223–224 Language Assessment Battery (LAB), 34, 35
instruction in, 126, 127–129, 141–142 Language Assessment Scales (LAS), 34, 35
literacy in, L2 literacy development and, Language Assessment Scales-Oral (LAS-O),
77–78 19
maintenance of, class and, 33–34 language minority students, 1
oral, proficiency in, literacy in L2 and, direct instruction of, 111
67–69 interactive instruction for, 116–117
studies of oral proficiency in and L2 Lara-Alecio, R., 39
literacy, 88t–95t Larson, K. A., 189
use in home of, 134–136 “late-exit” programs, 5, 183–185
“L2” (second language), 11 learning disabilities, students with, 128
concepts lexicalized in, 38 future research on, 230
correlation between oral and reading literacy practices of parents of, 134
measures of, 18t–19t learning strategies, for oral-language
crosslinguistic relationships between L1 development, 19–21, 39–40
and, 64–65 Lee, O., 198
developmental interdependence Leopold, W., 227
hypothesis on relationship between L1 Lerner, S., 189, 193
and, 65 Levine, 79
developmental links between L1 and, 224 Li, D., 120
language use and oral development of, “limited English proficient” (LEP) students,
27–31 11
learning strategies for, 19–21 Lindholm, Kathryn J., 16, 70–71, 180, 186,
literacy development in, L1 literacy and, 189
77–78 Lindholm-Leary, K. J., 189, 190, 191–192,
literacy in, oral proficiency and, 69–72 193, 194
literacy in, oral proficiency in L1 and, Linn, R. L., 197–198
67–69 literacy
literacy strategies for, 79–80 assessment of instruction in, 136–139
oral language development for, 15–19, cognate vocabulary in, 75–76
40–41 family and, 133–134
phonological awareness in, 73 family and community in development
rates of oral language proficiency of, 131–132
development for, 26–27 future research on, 229
studies of component skills of L2 literary genre and types of, 80–81
development, 96t–100t instruction in, 109
studies of L2 literacy, 101t–108t interactive instruction for, 116–117
studies of L2 oral proficiency and L2 in L1, literacy development in L2 and,
literacy, 92t–95t 77–78, 83
studies of literacy in, L1 oral proficiency in L2, oral proficiency in L1 and, 67–69
and, 88t–95t in L2, oral proficiency in L2 and, 69–72
use outside of school of, 32 orthographic knowledge in, 74–75
Lambert, W. E., 33–34 phonological awareness in, 73–74
Lanauze, M., 64, 69, 78 socioeconomic status and development
Langer, J. A., 64, 78, 80 of, 132–133
language strategies for, 79–80
academic achievement influenced by, studies of component skills of L2 literary
188–190 development, 96t–100t
of instruction, 126–131 studies of L1 oral proficiency and L2
of instruction, studies of, 166t–170t literacy, 88t–95t
process approach to, 121 studies of L2 literacy, 101t–108t
P1: irk
Index 243
studies of L2 oral proficiency and L2 future research on, 229, 230

literacy, 92t–95t L1 proficiency in, literacy in L2 and, 67–69
literature, in process-based instruction, 121 L2 proficiency in, literacy in L2 and, 69–72
Loban, W., 42 language use and development of, 27–31
López, I. C., 190 learning strategies for, 19–21
Lord, C., 197, 199 longitudinal and cross-sectional studies
of, 24t–25t
MacSwan, J., 36 mathematics achievement influenced by,
mainstream education, 178, 190 188
Malave, L., 27 non-school factors in learning, 31–34
Maldonado, J. A., 128–129 personality and social factors in learning,
materials, instructional, 192 21–23
mathematics achievement, 179, 184, 185 rates of proficiency in development of,
influence of oral language on, 188 23–27, 40
Mayer, R., 75 relationships between written language
McEvoy, R., 137 development and, 64–65
McLaughlin, B., 111–112, 114 school factors in learning, 23–27
Merino, B., 35 studies of L1 proficiency in and L2
meta-cognitive learning strategies, 20–21 literacy, 88t–95t
Milk, R. D., 27 studies of L2 oral proficiency and L2
Minaya-Rowe, Liliana, x literacy, 92t–95t
Minicucci, C., 191 Orr, J. E., 197
Miramontes, O., 38, 79–80, 137–138 orthographic knowledge, 74–75, 76
Miyoshi, J., 199
Molinet-Molina, M., 37 Padron, Y. N., 110, 112–113
Mortensen, E., 128 parents
level of education of, 193
Nagy, W., 73, 74, 75 literacy practices of, 134
natural communicative language (NCL), 22 socioeconomic status of, 132
Nes, S., 120 Park, H. S., 68, 132, 135
newcomer programs, 5 Parker, R., 39
Nguyen, A., 77 Patrick, S., 28
Nielsen, F., 188–189, 193, 194 Pearson, B., 37
No Child Left Behind Act (U.S., 2001), 3, Pearson, P. D., 75
196 Pease-Alvarez, L., 27, 31, 33
definitions used in, 11 Peck, S., 28–29
non-school factors peer interaction studies, 28
in learning oral language, 31–34, 41 Peregoy, S., 71
studies of, 31–59t Perez, E., 71
use of L1 outside school as, 68 personality factors, in learning oral
language, 21–23, 40
O’Brien, Gisela, x phonological awareness, 73–74, 82
Ochoa, S. H., 39 Piersel, W. C., 197
O’Connor, C., 195 Pilkington, C. L., 196
Oller, D. K., 31, 32, 33, 37 Piña, M., 129, 179
O’Malley, J. M., 21, 198 Platt, E., 28
oral language, 14–15 Politzer, R., 130
assessment of proficiency in, 34–39, 41–42 Ponterotto, J. G., 197
contexts of, 65–66 positive and negative transfer, 67
correlation between reading measures Prater, D. L., 80, 114, 116
and, 18t–19t prior knowledge, 195–196
development of, 15–19 Pritchard, R. H., 67, 75, 132
P1: irk
244 Index
process-based instruction, 110, 121–124, science, achievement in, 195–196

139–140 Sentence Verification Technique (SVT), 198
research on, 162t–165t Shavelson, R. J., 226, 227
Pucci, S. L., 112, 133 Sheltered Instruction, 6, 115, 140
Shepard, L., 200
question forms, 16, 39 Shin, F., 77
Short, D. J., 115, 140, 191
Ramirez, D., 130 Short, Deborah, x
Ramirez, J. D., 181–182, 183, 184, 191 Shriberg, E., 16
Ramirez, Mary, x Silva, C., 124
random assignment, 130–131, 178, 202 SIOP (Sheltered Instructional Observation
reading, see literacy Protocol), 191, 231
direct instruction in, 111, 112, 114 Smith, M., 195
phonological awareness in, 73–74 Snow, C. E., 16–17, 19, 64, 69, 78
studies of achievement in, 179 social class
reading achievement maintenance of L1 language and, 33–34
literacy-based abilities used in, 66 see also socioeconomic status
oral proficiency correlated with, 17–19 social factors, in learning oral language,
reading comprehension, 32–33 21–23, 40
reciprocal interaction model, 191 socioeconomic status (SES), 132
Reese, L., 68, 70, 78, 81, 132, 136 academic achievement and, 193–194
Reis, Noni, x interactive instruction and, 120
residence in U.S., length of, 194–195 literacy development and, 132–133
Riches, Caroline, x sounds, phonological awareness in, 73
Rivas, G. A., 35 Spanish (language)
Rivera, Charlene, x cognates for English and, 75
Roberts, T., 74 dialects of, 37
Rochin, R. I., 193 studies of achievement in, 181
Rodriguea-Brown, F. V., 16 Sparim, G., 74
Rodriquez, S., 118, 119 spelling, 74–75, 83
Rolstad, K., 36 Spencer, M. L., 35
Roser, N., 121 Spharim, G., 17
Rottenberg, L., 28 standardized tests
Rousseau, M. K., 113–114 academic achievement measured by, 176,
Royer, J. M., 71, 78, 198 197
Rubison, R. M., 115 accommodations for, 198–200
Ruiz-de-Velasco, J., 2 cultural bias in, 138
Rumbaut, R. G., 132, 136, 138 decisions based on results of, 196
Rumberger, R. W., 189 ELLs exempt from, 3
language used at home and scores on, 68
Sadoski, M., 112 under No Child Left Behind Act, 3
Saladate, M., 128 proficiency levels established for, 198
Saunders, William, 119–120, 141, 229 socioeconomic status correlated with,
Saville-Troike, M., 20, 29, 30–31, 69, 132
71 used to evaluate language of instruction,
Schell, L. M., 115 127
Schon, I., 123, 124 validity of, 137
school factors see also assessment
in learning oral language, 23–31, 41 Steinberg, L., 230
studies of, 52t–58t Stevens, R. A., 197, 198
Schrank, F., 35, 36–37 Stoops Verplaetse, L., 191
Schumm, J. S., 134 Strong, M., 22–23
P1: irk
Index 245
Structured English Immersion (SEI), 5 Vázquez, L. A., 190

bilingual instruction and ESL compared vocabulary
with, 181–183 cognate vocabulary, 75–76
Syvanen, C., 118 direct instruction in, 111–112
language use outside of school and, 32
Tam, B. K. Y., 113–114 reading comprehension and, 71
Tang, Sau-Lim, x vocabulary development, 16–17
Taylor, D. M., 33–34 Vogt, M. E., 115, 140
Taylor, G., 200 Vojir, C., 123, 124
teachers, 187 Vygotsky, L., 116
teaching, models of, 191
technology, in teaching, 192, 204 Ward, W., 190
Tellez, K., 187, 202 Weffer, R. E., 136, 194, 195
Terrasi, S., 74 Wen-Hui, L., 133
Tharp, Roland, x Wheeler, E., 28
Thomas, W. P., 129, 132, 179, 180, 182, 183 Willig, A. C., 202
threshold hypothesis, 66, 70 Winsler, A., 27
time, in learning, 225–226 Wong-Fillmore, 22
Tompkins, G. E., 67, 75, 132 Woodcock Language Proficiency Battery
Towne, L., 226, 227 (WLPB), 19
transitional bilingual programs, 5, 128 Woodcock Language Proficiency
Troudi, S., 28 Battery-Revised (WLPB-R), 26, 35
two-way (dual language) immersion Woodward, J., 128, 180, 182–183, 201
programs, 3, 25, 129, 179 writing
descriptive studies of, 185–220t direct instruction in, 111, 114
process approach to learning, 123
Ulanoff, S. H., 112, 133 written language
Ulibarri, D. M., 35, 36–37 oral proficiency in L1 and, 69
Umbel, V. M., 31, 32, 33, 37, 138 relationships between oral language
U.S.-born English language learners, 2 development and, 64–65
spelling in, 74–75
Valdez-Pierce, L., 198 see also literacy
Vasquez, O., 64
Vaughn, S., 110, 134, 192 Zutell, J., 75

Educating English Language Learners

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Educating English Language Learners

Uploaded by

Copyright:

Available Formats

P1: irk

0521859751pre CB989B/Genesee 0 521 85975 1 November 12, 2005 17:3

This page intentionally left blank

Educating English Language Learners: A Synthesis

Fred Genesee is Professor in the Department of Psychology, McGill University, Montreal.

Kathryn Lindholm-Leary is Professor of Child and Adolescent Development at San Jose

Educating English Language Learners

Cambridge University Press

© Cambridge University Press 2006

This publication is in copyright. Subject to statutory exception and to the provision of

First published in print format

isbn-13 978-0-511-14648-0 eBook (EBL)

isbn-13 978-0-521-85975-2 hardback

List of Tables page vii

Appendix A: Definitions of Abbreviations in Research Summary Tables 235

1.1. Characteristics of Program Alternatives for English

recommendations for future research agendas. The chapters in this volume

This volume synthesizes research on the relationships among oral lan-

increased by only 12 percent. In 2001–02, over 4.7 million school-aged chil-

The increasing number of students for whom English is an additional

program alternatives for ells

proficiency proficiency proficiency proficiency academic

proficiency, academic development, and crosscultural understanding. All

The core curriculum includes English language development (ELD), and

methodology for the review

We convened a thirteen-member team of researchers knowledgeable

Phase 1: Inclusion Parameters

Phase 2: Literature Searches and Quality Indicators

the research. The subgroups conducted several types of searches. First, we

Phase 3: Synthesis of Research

the research base. Thus, we developed a framework for characterizing the

organization of the volume

Federal Research Centers Technical Reports Searched

William M. Saunders and Gisela O’Brien

Specific Features of L2 Oral Language Development

16 William M. Saunders and Gisela O’Brien

of research of this sort, two domains of L2 oral language acquisition have

the source for the examples listed previously) is particularly relevant

table 2.1. Correlations Between L2 Oral and L2 Reading Measures

Formal Definitions 0.33a

Language Learning Strategies

20 William M. Saunders and Gisela O’Brien

Chesterfield and Chesterfield found a hierarchical relationship among

tasks; discrete tasks included practicing pronouncing words and learning

Personality and Social Factors

22 William M. Saunders and Gisela O’Brien

playmate nominations was plotted across the year, as students’ English

Rates of Oral Language Proficiency Development

interpreted as native-like proficiency, and Level 4 is viewed as sufficient for

26 William M. Saunders and Gisela O’Brien

Language Use and L2 Oral Development

28 William M. Saunders and Gisela O’Brien

30 William M. Saunders and Gisela O’Brien

In contrast, among pre-school ELLs and over a longer duration of time,

communication with each other and with native English-speaking peers, as

32 William M. Saunders and Gisela O’Brien

L2 reading comprehension. Hansen concluded that while home language

34 William M. Saunders and Gisela O’Brien

36 William M. Saunders and Gisela O’Brien

levels of proficiency (e.g., non-proficient, limited-proficient, fluent-English

(r = 0.80), and IPT-I, typically viewed as less academically oriented, yielded

38 William M. Saunders and Gisela O’Brien

language without some knowledge about the composition of the norm

40 William M. Saunders and Gisela O’Brien